Ionised and total hypocalcaemia in pregnancy: An analysis of prevalence and risk factors in a resource-limited setting, Cameroon.

INTRODUCTION: Hypocalcaemia remains a prevalent laboratory finding in pregnancy, capable of inducing adverse maternofoetal outcomes. This study compares the prevalence of hypocalcaemia in apparently healthy pregnant women from the ionised, and total calcaemia viewpoints and further identifies factors associated with total crude and ionised hypocalcaemia in pregnancy.
METHODS: A hospital-based cross-sectional study was conducted between November 2020 and September 2021, targeting apparently healthy pregnant women received in late pregnancy in four maternities in the Nkongsamba Health District, Cameroon. Blood samples were collected and analysed for serum ionised calcium concentrations and pH (by ion-selective electrode potentiometry), and for total calcium and albumin concentration (by atomic absorption spectrophotometry). Sociodemographic, obstetric and nutritional data were collected using an interviewer-administered questionnaire.
RESULTS: The average age of the 1074 participants included in the study was 28.20±6.08 years. The prevalence of total crude and total albumin-corrected hypocalcaemia in this study was 61.64 [58.69-64.50]% and 56.70 [53.72-59.64]%, respectively (p-value = 0.000). The prevalence of ionised hypocalcaemia was very low (2.89 [2.04-4.07]%) compared with the prevalence of total hypocalcaemia (p-value = 0.000). Monthly income below 100.000FCFA (179 USD) (AOR = 0.73, p-value = 0.024), taking more than 2 meals daily (AOR = 0.68, p-value = 0.006) and taking desserts (AOR = 0.73, p-value = 0.046) reduced the odds of total crude hypocalcaemia, while having banana/plantain and tubers as the content of their most consumed meal significantly increased the odds of total crude hypocalcaemia (AOR = 1.37, p-value = 0.012). Single women (AOR = 2.54, p-value = 0.021), with a higher education (AOR = 3.27, p-value = 0.017), who initiated antenatal care before 4 months (AOR = 2.47, p-value = 0.029), had their odds of ionised hypocalcaemia significantly increased. On the other hand, women below 30 years (AOR = 0.44, p-value = 0.039), with occupations other than housewife (AOR = 0.34, p-value = 0.027), and women who took desserts between meals (AOR = 0.45, p-value = 0.034) had their odds of ionised hypocalcaemia significantly reduced. Taking calcium supplements simultaneously with other supplements also significantly reduced the odds of total hypocalcaemia in pregnancy (OR = 0.69, p-value = 0.027).
CONCLUSION: Ionised hypocalcaemia in pregnancy is a rare finding. Only 2.89% of all apparently healthy pregnant women have ionised hypocalcaemia in late pregnancy, while 56.70% have total hypocalcaemia. Factors like the daily number of meals, taking of desserts, the content of the most consumed meal and monthly revenue significantly affect the prevalence of total hypocalcaemia in pregnancy. On the other hand, factors like age above 30 years, having a higher education, being single, having initiated antenatal care before 4 months of pregnancy, being a housewife and not taking desserts between meals have a significantly positive association with ionised hypocalcaemia.

Introduction

Hypocalcaemia in pregnancy remains a prevalent finding combatted by the current World Health Organisation (WHO) recommendations on calcium supplementation. According to evidence from cross-sectional surveys, the prevalence of total hypocalcaemia in the third trimester of pregnancy is very high, with variations depending on the setting. In India and Algeria, the third-trimester prevalence of total hypocalcaemia was 66% [1] and 70% [2], respectively. In Cameroon, a recent study at the Nkongsamba Regional Hospital (NRH) reported a total hypocalcaemia prevalence of 59% [3]. In Nigeria, the prevalence of total hypocalcaemia in pregnancy (including women from all trimesters, with a higher cut-off of 88mg/L) was still as high as 29% [4]. According to a small-scale Nigerian study, this prevalence of total hypocalcaemia increases with gestational age, passing from 25.6% in the first two trimesters to 40% in the third trimester [5]. All the above studies measured total calcaemia and not ionised fractions of calcium in blood.

It is argued that total hypocalcaemia in pregnancy, even though a common finding, can be ignored because ionised calcium concentrations remain sensibly within normal range throughout pregnancy [6]. This has discouraged studies aimed at measuring the burden of ionised hypocalcaemia in pregnancy. Notwithstanding, growing evidence is associating hypocalcaemia in pregnancy and low calcium intake to hypertensive diseases in pregnancy, a leading cause of maternal mortality worldwide. Low calcium intake increases the likelihood of foetal growth restriction, foetal low bone mass, low birth-weight and shorter birth length [6].

Insufficient calcium intake before and during pregnancy has been reported in systematic reviews to characterise developing countries like Cameroon [7, 8]. The high prevalence of hypocalcaemia in Cameroon could be associated with nutritional factors and failure to supplement. Despite recommendations from the WHO for systematic supplementation with calcium during pregnancy in regions with insufficient calcium intake (principally to prevent hypertensive diseases in pregnancy and their complications) [9], the recent study carried out in Cameroon found out that 43% of women went through pregnancy without taking calcium supplements [3].

Despite the risk associated with hypocalcaemia in pregnancy, very few studies have tried to identify the risk factors associated with this imbalance in pregnancy. In one Malaysian study, factors like low milk intake in pregnancy, underweight and obesity before pregnancy werefound to be associated with an increased likelihood of hypocalcaemia in pregnancy [10]. Another study in Nigeria found that primigravidity, anaemia in pregnancy, anorexia and a low level of education significantly increased the occurrence of hypocalcaemia in pregnancy [4].

Apart from calcium supplementation, a great deal of calcium requirements are met in our diets [8, 11]. Due to food unavailability, inaccessibility, and wastage, Cameroon is classified as a country with food insecurity. According to the World Food program, 16% of Cameroonian households suffer from food insecurity [12]. Results from a study in Yaoundé, Cameroon, show that only 38% of women knew they should have at least 3 meals a day while only 22% actually practiced this [13]. Fruit and vegetable consumption have been reported to have beneficial effects on calcium stores and bone mineral density [14, 15]. As far as fruit consumption and availability is concerned, pineapples, watermelons, and oranges are classified to be most accessible and commonly consumed [12].

To the best of our knowledge no study in Cameroon has been designed to measure the burden of ionised hypocalcaemia in pregnancy, and very little interest has been put on sociodemographic, nutritional and obstetric factors affecting calcaemic states in pregnancy. Moreover, sociodemographic, nutritional and obstetric factors associated with hypocalcaemia in pregnancy are still understudied in Sub-Saharan Africa. Globally, no study has measured the prevalence of hypocalcaemia in pregnancy from the ionised calcium viewpoint. This work compares the prevalence of ionised hypocalcaemia in apparently healthy pregnant women to the total hypocalcaemia prevalence. It also presents the sociodemographic, obstetrical and nutritional factors associated with ionised and total hypocalcaemia in pregnancy among a sample of pregnant women in the Nkongsamba Health District (NHD), Cameroon.

Methodology

The methods approved for obtaining the data have been published in PLoS One [16]. They have been summarised below with highlights of the sections used to obtain data for this work.

Study design and setting

We conducted a hospital-based cross-sectional study targeting women in late pregnancy from four major health facilities of the Nkongsamba Health District (NHD). Contrary to the initial indications on the protocol that data were to be collected only from the NRH, 4 major health facilities of the NHD were included in the study to work with a more representative population. These health facilities included the NRH, Catholic Medicalised Health Centre, Fultang polyclinic and the Bon Samaritain Medicalised Health Centre. These health facilities are responsible for over 85% of deliveries that occur in the health district.

Data were collected using a semi-structured interview-administered questionnaire and by blood assays. All apparently healthy women were involved in the study. Data collection for this paper was done between November 2020 and September 2021.

Target population and sampling

Eligible participants were all pregnant women received at the antenatal care unit of the maternity of the four health facilities in late pregnancy (at least at 37 weeks of gestation) [16]. Contrary to the information presented on the protocol, which planned to recruit only apparently healthy women with no potential causes of hypocalcaemia, this project recruited all apparently healthy pregnant women to report a general prevalence of hypocalcaemia. The minimum required sample size stated in the protocol was 1067 participants, and participants were consecutively included [16] in the study as they arrived at the maternity of these health facilities. Recruitment stopped when the required sample size was reached, and non-response cases were replaced.

Procedure of implementation and data collection

The research questionnaire was pretested at the Kekem District Hospital, West of Cameroon, while administrative authorisations were obtained from the District Medical Officer of the NHD and the directors of the respective health facilities. Data collection was done under supervision by seven midwives working in the included maternities. These data collectors were trained in three training sessions, which lasted five hours each on the consenting and data collection procedures.

Only eligible and consenting pregnant women were allowed to participate in this study. Questionnaires were administered face-to-face, on a one-to-one basis, and data gathered included sociodemographic, economic, and obstetric characteristics. In addition to the interview, 10ml of total blood were collected into lithium heparinised tubes following the WHO best practices in phlebotomy using vacuated needles on unsqueezed veins to measure total calcaemia. These samples were immediately centrifuged and used to measure total calcaemia and albuminaemia. Moreover, 10ml of blood was collected into dry vacutainer tubes (without anti-coagulant) and settled for 30 minutes for serum extraction (immediately used for ionised calcium and pH assays). All blood samples were collected on seated participants after breathing calmly for 10 minutes, with no prior physical exercise. During this procedure, tourniquets were applied only for less than 1 minute, arm exercise and fist formation were discouraged [16].

Payne’s equation [17, 18] is a valid correction equation when albumin concentrations are measured using Bromocresol Green (BCG) [19]. However, at very low albumin concentrations, normocalcaemic cases risk to be classified to be hypercalcaemic [20]. According to a recent analysis of existing evidence, total calcium levels should be used instead of the albumin-corrected calcium in patients with no suspected disorder of calcium homeostasis [21]. Given the recent controversy on this subject and no established consensus in pregnancy, our findings were presented using the total and albumin-corrected calcium concentrations.

Total blood calcium and albumin were measured by atomic absorption spectrophotometry (AAS) using the semi-automatic KENZA MAX spectrophotometer manufactured by BIOLABO. Total blood calcium and albumin were measured using the O-Cresol Phtalein Complexone (CPC) and bromocresol green (BCG) methods, respectively (as per the BIOLABO standard operating procedures) [22, 23]. Crude ionised calcium concentrations and pH were measured by ion-selective electrode potentiometry (ISEP) using the K-lite 8 (MSLEA15-H model) serum electrolyte analyser and its standard reagents following standard manufacturer protocols [3]. Crude ionised calcium concentrations were then corrected for any pH variations around 7.4 using the equation Corrected iCa2+ (pH 7.4) = Measured iCa2+ [1–0.53 × (7.40–measured pH)] [24] which is valid between 7.2–7.6.

Data analysis

A data entry sheet was developed on Epi-Info version 7.2.4.0, and all validated questionnaires were entered into the sheet. Statistical analysis was carried out using Epi-Info version 7.2.4.0. Questionnaires were evaluated for completeness and the absence of key information like the albuminemia, crude ionised and total calcaemia. When vital information key to the study was missing, the questionnaire was rejected. Major frequencies or proportions were estimated with their 95% confidence intervals (CIs) for categorical variables (prevalence of hypocalcaemia, marital status, religion, and occupation), while the mean (μ) and standard deviation (SD) were used for continuous variables (age, calcaemia, and albuminemia). Participants with total crude calcium concentrations or total albumin-corrected calcium concentrations less than or equal to 85mg/L [3, 6] were considered to be hypocalcaemic. In the case of ionised hypocalcaemia, a cut-off of less than or equal to 45mg/L was used [25]. All women who had ionised, total crude, and albumin-corrected total calcium concentrations less than or equal to the stated cut-offs were classified as hypocalcaemic (low blood calcium).

Concerning logistic regression, total crude hypocalcaemia (total calcaemia measured by AAS and hypocalcaemia defined with a cut-off of 85mg/L) and ionised hypocalcaemia (ionised calcaemia measured by ISEP, and hypocalcaemia defined with a cut-off of 45mg/L) were used as the outcomes of interest. Because of the absence of identical studies, most of the categorisation of sociodemographic characteristics for logistic regression was done following directions of effect [3, 26] and examples from previous studies among pregnant women [27]. Other variables like the marital status were categorised following standard divisions like in union/out of union [3]. Obstetric factor categorisation was done based on evidence from other studies [28-30] and WHO indicators antenatal care coverage [31, 32]. People who consume fruits rarely are more exposed to hypocalcaemia than their counterparts who take fruits more often [14]. So, women who took fruits rarely (less than 2 times a week) were compared with those who took fruits more often. The strength of association between hypocalcaemia (ionised and total) and selected covariates like monthly revenue, level of education, number of antenatal visits, and nutritional habits was measured using the odds ratio (OR), and 95% CI generated during logistic regression. Variables with p-values < 0.25 were considered for multiple logistic regression. In a multiple logistic regression model, each eligible variable from simple logistic regression was controlled for all other eligible variables. All the variables eligible for multiple logistic regression remained in the model with no backward elimination. The threshold of significance for this analysis was set at a p-value of 0.05.

Ethics statement

Ethical clearance was obtained from Cameroon Bioethics Initiative (CAMBIN)/ Ethics Review and Consultancy Committee (ERCC). The ethical clearance reference number is CBI/452/ ERCC/CAMBIN. Only consenting participants who signed informed consent forms (or assent forms for minors) were included in this study. The confidentiality and autonomy of all participants were respected, and they were free to leave the study at any time.

Results

Characteristics of the population

This study included a total of 1074 eligible individuals with a non-response rate of 6.61%. The mean age of the participants was 28.20±6.08 years, with the 21–30 years age group being the most represented (56.52%). Table 1 shows the sociodemographic characteristics of the participants. About 9 in 10 (93%) had attended secondary school, and 65% were in a union. Concerning religion and monthly revenue, 90.81% were Christians and 11.24% earned below 50.000FCFA (89 USD) every month.

Table 1

Sociodemographic characteristics of participants.

Characteristic	Modalities	Frequency	Proportions (%)
Age groups(n = 1074)	15–20 years	97	09.03
	21–30 years	607	56.52
	31–50 years	370	34.45
Marital status(n = 1068)	Single	371	34.74
	Married	269	25.19
	Cohabiting	462	39.89
	Widow	02	0.19
Level of education(n = 1074)	Primary	72	06.60
	Secondary	741	68.99
	Higher	261	24.30
Monthly revenue in thousand FCFA(n = 1059)	< 50 (USD 89)	119	11.24
	50–100 (USD 89–179)	270	25.50
	100–200 (USD 179–358)	456	43.06
	> 200 (USD 158)	214	20.21
Religion(n = 1056)	Atheist	64	06.06
	Catholic	518	49.05
	Muslim	33	03.13
	Protestant	441	41.76
Number of household occupants(n = 1053)	1–4	438	41.60
	5–7	415	39.41
	≥ 7	200	18.99

The obstetric characteristics of the participants are presented in Table 2. The mean number of pregnancies and the average gestational age at the first antenatal visit (for the current pregnancy) was 3.46±2.06 and 17.10±8.08 weeks, respectively. About 2 in 10 women were in their first pregnancy, while a majority of the participants were in their second, third or fourth pregnancy (47.06%). About half (53.52%) initiated antenatal care after 4–6 months of pregnancy, while only about 1 in 10 women (11.62%) started antenatal care within the first two months of pregnancy. Concerning antenatal visits, 87.14% had attended at least 4 antenatal care sessions.

Table 2

Obstetric characteristics of participants.

Characteristic	Modalities	Frequency	Proportion (%)
Gestational age at first antenatal visit(n = 1050)	Before 2 months	122	11.62
	2–3 months	263	25.05
	4–6 months	562	53.52
	Above 7 months	103	09.81
Number of antenatal visits(n = 1034)	No antenatal care	09	0.87
	1–3	124	11.99
	4–6	554	53.58
	> 6	347	33.56
Total number of pregnancies(n = 1071)	First pregnancy	256	23.90
	2–4	504	47.06
	5–8	289	26.98
	> 8	22	02.05
Total number of deliveries(n = 1067)	None	249	23.34
	1–3	596	55.86
	4–6	205	19.21
	> 6	17	01.59

Prevalence of hypocalcaemia in late pregnancy

In this study, the prevalence of total crude and albumin-corrected hypocalcaemia were 61.64 [58.69–64.50]% and 56.70 [53.72–59.64]%, respectively. The difference in the above two proportions was statistically significant with a p-value of 0.000. From the pH-corrected ionised calcium point of view, the prevalence of ionised hypocalcaemia was 2.89 [2.04–4.07]%. When compared with the total crude and total albumin-corrected prevalence of hypocalcaemia, this showed statistically significant differences with p-values of 0.000 in each case.

Only 28 participants (2.61 [1.81–3.74]%) had hypocalcaemia when defined using the three methods (ionised, total crude and albumin corrected total calcium) at the same time. Fig 1 shows the scatter plot of the relationship between ionised calcaemia and total crude calcaemia, while in Fig 2, ionised calcaemia is plotted against albumin-corrected total calcium. The two plots show that trivial increases in ionised calcaemia are associated with more considerable changes in total crude and albumin-corrected calcaemia. The goodness-of-fit measure (R2) in both plots shows that the strength of the relationship is very weak, with R2 less than 5% in both cases. The probability of having the gradient zero in each case was 0.000.

Fig 1

Plot of the relationship between ionised calcaemia and total crude calcaemia in pregnancy.

Fig 2

Plot of the relationship between ionised calcaemia and albumin-corrected total calcaemia in pregnancy.

The mean concentrations of total albumin-corrected and total crude calcaemia (uncorrected calcaemia) were 82.86±8.89 mg/L and 82.20±8.84 mg/L, respectively. The mean crude and pH-corrected ionised calcaemia among these 1074 participants were 52.87±3.83 mg/L and 52.61±3.07 mg/L, respectively. Women classified to have total crude hypocalcaemia had their odds of ionised hypocalcaemia increased by 5.88 folds compared with their counterparts with total normocalcaemia (p-value = 0.003).

Factors associated with total crude hypocalcaemia in pregnancy

Sociodemographic and obstetric factors associated with total hypocalcaemia are presented in Table 3. Following simple logistic regression, only monthly revenue and religion were significantly associated with hypocalcaemia. When controlled in a multiple logistic regression model with six eligible factors (p-value < 0.25), only monthly revenue showed significant association with total crude hypocalcaemia. Participants who earned less than 100.000FCFA (179 USD) in a month had 0.73 lower odds to have total crude hypocalcaemia than those who earned more. No factors related to antenatal care were found to be statistically significant.

Table 3

Sociodemographic and obstetric factors associated with total crude hypocalcaemia.

Factor	Simple logistic regression		Multiple logistic regression
	OR [95% CI]	p-value	AOR [95% CI]	p-value
Age below 30 (Y/N)	0.93 [0.72–1.21]	0.604
Level of education of participant above secondary (Y/N)	0.86 [0.65–1.15]	0.315
Level of education of partner above secondary (Y/N)	0.79 [0.60–1.03]	0.079	0.81 [0.60–1.09]	0.163
Less than 7 household occupants (Y/N)	1.19 [0.91–1.56]	0.198	1.13 [0.85–1.51]	0.394
Marital status as single (Y/N)	0.90 [0.69–1.16]	0.416
Gestational age at first visit below 4 months (Y/N)	1.05 [0.81–1.36]	0.707
Number of ANC visits above 3 (Y/N)	0.97 [0.67–1.36]	0.878
Occupation other than housewife (Y/N)	0.80 [0.62–1.04]	0.101	0.94 [0.70–1.25]	0.669
Monthly revenue below 100.000 FCFA (Y/N)	0.75 [0.58–0.97]	0.027	0.73 [0.55–0.96]	0.024
Non-Catholics (Y/N)	1.30 [1.01–1.67]	0.039	1.23 [0.95–1.59]	0.120
Always or sometimes reminded or supported by partner to take supplements (Y/N)	0.79 [0.56–1.11]	0.183	0.74 [0.52–1.07]	0.115

• p-value < 0.05: Statistical significance,

• p-value < 0.25: Eligible for multiple regression analysis,

• All the variables eligible for multiple logistic regression remained in the model with no backward elimination,

• Total hypocalcaemia: Total crude calcium concentration less than or equal to 85.0 mg/L (Measured by atomic AAS using the O-Cresol Phtalein Complexone reagent),

• Prevalence of total crude hypocalcaemia (662/1074) = 61.64%,

• Y/N = Yes/No, OR = Odds Ratio, AOR = Adjusted Odds Ratio, CI = Confidence Interval.

Table 4 presents nutritional factors associated with total crude hypocalcaemia. Following simple logistic regression, taking more than two meals a day in pregnancy, taking of desserts, cheese consumption, and consumption of cereals like rice and corn (as the most common food) were significantly associated with lower odds of total crude hypocalcaemia in pregnancy. When controlled for confounders in a multiple logistic regression model, women who took more than 2 meals a day (AOR = 0.68 [0.52–0.90], p-value = 0.006) and those who took desserts (AOR = 0.73 [0.54–0.99], p-value = 0.046) had their odds of total crude hypocalcaemia reduced by 0.68 and 0.74 fold, respectively. Moreover, having banana/plantain and tubers as the content of their most consumed meal (as opposed to cereals) significantly increased the odds of total crude hypocalcaemia by 1.37 folds (AOR = 1.37 [1.08–1.80], p-value = 0.012).

Table 4

Nutritional factors associated with total crude hypocalcaemia.

Factor	Simple Logistic regression		Multiple logistic regression
	OR [95% CI]	p-value	AOR [95% CI]	p-value
Daily number of meals above 2 (Y/N)	0.68 [0.52–0.88]	0.004	0.68 [0.52–0.90]	0.006
Taking desserts between meals (Y/N)	0.74 [0.55–0.99]	0.043	0.73 [0.54–0.99]	0.046
Consumption of plantain, banana and tubers compared with Others (like corn and rice) as most common content of the meal	1.36 [1.06–1.75]	0.015	1.37 [1.08–1.80]	0.012
Taking fruits rarely (Y/N)	0.82 [0.50–1.34]	0.434
Taking oranges, watermelon, and pineapples as the most consumed fruit (Y/N)	1.06 [0.81–1.37]	0.689
Never taken cheese (Y/N)	1.25 [0.97–1.60]	0.080	1.11 [0.85–1.45]	0.430

• p-value < 0.05: Statistical significance,

• p-value < 0.25: Eligible for multiple regression analysis,

• All the variables eligible for multiple logistic regression remained in the model with no backward elimination,

• Total hypocalcaemia: Total crude calcium concentrations less than or equal to 85.0 mg/L (Measured by AAS using the O-Cresol Phtalein Complexone reagent),

• Prevalence of total crude hypocalcaemia (662/1074) = 61.64%,

• Y/N = Yes/No, OR = Odds Ratio, AOR = Adjusted Odds Ratio, CI = Confidence Interval.

Table 5 presents the sociodemographic and obstetric factors associated with ionised hypocalcaemia. Contrary to total hypocalcaemia, which had monthly revenue as the only sociodemographic factor, 5 factors were found to have a significant association with ionised hypocalcaemia. Age, level of education, marital status, gestational age at the first visit, and the participant’s occupation had a statistically significant association with ionised hypocalcaemia. Women below 30 years had 0.44 lower odds of ionised hypocalcaemia compared with their older counterparts (AOR = 0.44 [0.20–0.96], p-value = 0.039). Those with higher education had 3.27 higher odds of ionised hypocalcaemia compared with women with lower education (AOR = 3.27 [1.23–8.70], p-value = 0.017). Single women had their odds for ionised hypocalcaemia increased by 2.54 folds compared with other women in union (AOR = 2.54 [1.15–5.62], p-value = 0.021). Women who started ANC before 4 months of pregnancy had their odds of ionised hypocalcaemia increased by 2.47 folds compared with their counterparts who started later (AOR = 2.47 [1.10–5.56], p-value = 0.029). Moreover, women with occupations other than housewife had their odds of ionised hypocalcaemia reduced by 0.34 folds (AOR = 0.34 [0.13–0.89], p-value = 0.027).

Table 5

Sociodemographic and obstetric factors associated with ionised hypocalcaemia.

Factor	Simple logistic regression		Multiple logistic regression
	OR [95% CI]	p-value	AOR [95% CI]	p-value
Age below 30 (Y/N)	0.55 [0.27–1.13]	0.102	0.44 [0.20–0.96]	0.039
Level of education of participant above secondary (Y/N)	1.74 [0.82–3.69]	0.146	3.27 [1.23–8.70]	0.017
Level of education of partner above secondary (Y/N)	1.33 [0.53–2.43]	0.749
Less than 7 household occupants (Y/N)	0.92 [0.43–1.97]	0.828
Marital status as single (Y/N)	1.90 [0.92–3.92]	0.084	2.54 [1.15–5.62]	0.021
Gestational age at first visit below 4 months (Y/N)	2.01 [0.97–4.18]	0.059	2.47 [1.10–5.56]	0.029
Number of ANC visits above 3 (Y/N)	0.55 [0.22–1.39]	0.207	0.38 [0.13–1.11]	0.076
Occupation other than housewife (Y/N)	0.62 [0.30–1.26]	0.186	0.34 [0.13–0.89]	0.027
Monthly revenue below 100.000 FCFA (Y/N)	1.09 [0.52–2.27]	0.817
Non-Catholics (Y/N)	1.46 [0.70–3.06]	0.317
Always or sometimes reminded or supported by partner to take supplements (Y/N)	0.69 [0.27–1.72]	0.422

• p-value < 0.05: Statistical significance,

• p-value < 0.25: Eligible for multiple regression analysis,

• All the variables eligible for multiple logistic regression remained in the model with no backward elimination,

• Ionised hypocalcaemia: pH-corrected ionised calcium concentration less than or equal to 45.0 mg/L (Measured by ISEP),

• Prevalence of ionised hypocalcaemia (31/1074) = 2.89%,

• Y/N = Yes/No, OR = Odds Ratio, AOR = Adjusted Odds Ratio, CI = Confidence Interval.

Table 6 presents nutritional factors associated with ionised hypocalcaemia (only one variable was eligible for multiple logistic regression). Contrary to total hypocalcaemia, which was more associated with nutritional factors, ionised hypocalcaemia was associated only with the habit of taking desserts between meals. Women who took desserts between meals had their odds of ionised hypocalcaemia reduced by 0.45 folds (OR = 0.45 [0.21–0.94], p-value = 0.034).

Table 6

Nutritional factors associated with ionised hypocalcaemia in pregnancy.

Factors	Simple Logistic regression
	OR [95% CI]	p-value
Daily number of meals above 2 (Y/N)	0.71 [0.34–1.45]	0.344
Taking desserts between meals (Y/N)	0.45 [0.21–0.94]	0.034
Consumption of plantain, banana and tubers compared with Others (like corn and rice) as most common content of the meal	1.08 [0.52–2.25]	0.839
Taking fruits rarely (Y/N)	1.66 [0.48–5.69]	0.416
Taking oranges, watermelon, and pineapples as the most consumed fruit (Y/N)	0.72 [0.32–1.62]	0.420
Never taken cheese (Y/N)	1.48 [0.70–3.13]	0.299

• p-value < 0.05: Statistical significance

• Ionised hypocalcaemia: pH-corrected ionised calcium concentration less than or equal to 45.0 mg/L (Measured by ISEP).

• Prevalence of ionised hypocalcaemia (31/1074) = 2.89%,

• Y/N = Yes/No, OR = Odds Ratio, CI = Confidence Interval.

Table 7 evaluates calcium supplementation as a factor influencing calcaemic states. From the table, calcium supplemented women had respectively, 0.84 and 0.79 lower odds of total (p-value = 0.231) and ionised hypocalcaemia (p-value = 0.538) compared with non-supplemented women. Also, taking calcium supplements at the same time as other supplements (iron and folic acid) significantly reduced the odds of total crude hypocalcaemia in pregnancy (OR = 0.69 [0.50–0.96], p-value = 0.027). Taking more than 1000mg of elemental calcium a day was associated with reduced odds of total crude hypocalcaemia (OR = 0.75 [0.55–1.01], p-value = 0.061). No statistically significant association was found between ionised hypocalcaemia and calcium supplementation variables.

Table 7

Logistic regression analysis between calcium supplementation variables and likelihood of total crude and ionised hypocalcaemia.

	Total crude hypocalcaemia		Ionised hypocalcaemia
Calcium supplementation variable	OR [95% CI]	p-value	OR [95%CI]	p-value
Calcium supplementation (Y/N)	0.84 [0.63–1.12]	0.231	0.79 [037–1.69]	0.538
Duration of calcium supplementation below 4 months (Y/N)	0.89 [0.62–1.27]	0.522	0.40 [0.09–1.74]	0.224
Taking calcium supplements with other supplements (Y/N)	0.69 [0.50–0.96]	0.027	0.80 [0.32–2.03]	0.651
Daily dose of elemental calcium supplementation above 1000mg (Y/N)	0.75 [0.55–1.01]	0.061	2.32 [0.92–5.82]	0.073

• p-value < 0.05: Statistically significance,

• Total hypocalcaemia: Total crude calcium concentrations less than or equal to 85.0 mg/L (Measured by AAS using the O-Cresol Phtalein Complexone reagent),

• Ionised hypocalcaemia: pH-corrected ionised calcium concentration less than or equal to 45.0 mg/L (Measured by ISEP).

• Prevalence of total crude hypocalcaemia (662/1074) = 61.64%,

• Prevalence of ionised hypocalcaemia (31/1074) = 2.89%,

• Y/N = Yes/No, OR = Odds Ratio, CI = Confidence Interval.

Discussion

The main goal of our study was to estimate the prevalence of hypocalcaemia in apparently healthy pregnant women in the third trimester from the ionised and total calcaemia viewpoints. In this work, factors associated with ionised and total hypocalcaemia were also evaluated. The prevalence of total crude and total albumin-corrected hypocalcaemia in this study were 61.64 [58.69–64.50]% and 56.70 [53.72–59.64]%, respectively (p-value = 0.000), while the prevalence of ionised hypocalcaemia was as low as 2.89 [2.04–4.07]%.

It has been argued that total crude and albumin-corrected calcaemia in pregnancy have shortcomings in defining the exact physiological calcaemic states. Defining hypocalcaemia in pregnancy from the viewpoint of total calcaemia exaggerates the burden of this condition. The enormous discrepancy observed between the results of hypocalcaemia in pregnancy when defined from the total and ionised calcium point of view has been described in literature. The common laboratory finding of low circulating total calcium in pregnancy is far from directly defining the physiologically active concentration and is more of a factitious hypocalcaemia [6].

This discordance between total and ionised calcaemia in classifying calcaemic states has been described in other studies. In a study on outpatients suspected of bone or calcium metabolism disorders, there was 92% disagreement between ionised and total calcaemia in classifying hypocalcaemic states [33].

The scientific evidence suggesting that ionised calcium concentrations remain constant throughout pregnancy despite the variations in total calcium concentrations [6, 34] has limited exploration of the burden of ionised hypocalcaemia in pregnant women. Even if the state of pregnancy does not affect ionised calcaemic states, there are many potential causes of hypocalcaemia that can exist in association with pregnancy. These causes are dominated by vitamin D deficiencies, hypoparathyroidism, nutritional deficiencies and malnutrition, and some chronic diseases [6]. Therefore, ionised hypocalcaemia can be a problem in some pregnancies. To the best of our knowledge, no study has determined the prevalence of ionised hypocalcaemia in pregnancy.

This state of ionised hypocalcaemia is directly associated with adverse effects on maternal calcium-related physiology and the growth of the foetus [6]. Calcium has been identified to interfere in practically all physiological activities in the human body. It is indispensable as a second messenger in major signalling pathways and contributes to regulating cell excitability, exocytosis, apoptosis, and cell movement. Even though known to be governed mainly by the phospholipase C pathway, Ca2+ signalling can be engaged by many other pathways, including pathways for cell growth, differentiation and cell death. Calcium and calcium signalling are indispensable for vital activities like cell death, muscle contraction, neuronal transmission, neurogenesis, gene transcription, exocytosis, cell movement, cell growth and proliferation, synaptic plasticity, enzyme activity, secretion of saliva, and more [35, 36]. Low levels of ionised calcaemia in pregnancy are likely to interfere with most of these processes, foetal growth and differentiation-related activities.

The high prevalence of total hypocalcaemia and the cases of ionised hypocalcaemia in pregnancy in this population might be due to the high prevalence of vitamin D deficiency and insufficiency that have been reported in low and middle-income countries. A study in Buea (Cameroon) reported a prevalence of vitamin D deficiency and insufficiency in adults beyond 35 years of 3.2% and 22.6% [37]. In neighbouring Nigeria, the third-trimester prevalence of vitamin D deficiency and insufficiency of 22.5% and 60%, respectively, have been reported [38]. The prevalence of hypoparathyroidism in the general population is low, ranging from 0.5–6.6% [39], while the prevalence in pregnancy has not been reported [40].

The recorded prevalence of ionised hypocalcaemia in this study is likely due to subclinical hypovitaminosis D, hypoparathyroidism or other subclinical causes of hypocalcaemia in pregnancy. Therefore, apparently healthy pregnant women are likely to have potentially life-threatening pathologies liable to cause a fall in ionised calcaemia. In our study, this occurs in about 3 of every 100 pregnant women. This is not negligible, given that it could be associated with severe adverse maternofoetal morbidities. Pregnant women presenting with symptoms of hypocalcaemia should not only be placed on calcium therapy, but a complete exploration into the specific cause in pregnancy needs to be carried out.

As reported by studies in India, Algeria, and Cameroon, the prevalence of total hypocalcaemia in pregnancy is very high. These studies evaluated the prevalence of hypocalcaemia in the third trimester and found 66%, 70%, and 59% in India, Algeria, and Cameroon, respectively. The study in Cameroon was carried out in one of the health facilities considered in this study (NRH) and reported a 95% confidence interval of 53.42–63.90%, which perfectly intersects the confidence interval of the current research (56.70 [53.72–59.64]%).

According to the findings of this new study, at least 5 in 10 pregnant women in the NHD have low calcium concentrations in the third trimester. When total crude calcium concentrations are considered, the prevalence of hypocalcaemia increases slightly but significantly to 61.64 [58.69–64.50]% in this population. The slight discrepancy between these two results from Cameroon can be explained by an extension of the study population to include women from three other major health facilities in the NHD. Also, the prevalence of hypocalcaemia reported in India and Algeria are slightly higher than findings in Cameroon but similar to the prevalence recorded in this study with total crude (unadjusted or uncorrected) calcium concentrations (61.64%). These studies considered lower cut-offs (less than 80mg/L) to define hypocalcaemia, which might give us a lower prevalence if considered in this study. Moreover, Cameroon’s nutritional habits and sociodemographic characteristics are very different from those in India and Algeria (a possible explanation for the observed discrepancies).

However, a study in Maiduguri, Nigeria, reported a prevalence of hypocalcaemia in pregnancy of 29% (about half the prevalence in the present study) [4]. The inclusion criteria in this study could explain the low prevalence, given that they included pregnant women of all gestational ages [4]. Given that calcium absorption in pregnancy doubles within the first 12 weeks [34], some participants must have been recruited in the first and second trimesters when calcium concentrations are relatively high and foetal demands still low.

Should total hypocalcaemia in pregnancy be ignored, as suggested by some authors? Our study found out that total crude hypocalcaemia was significantly associated with increased odds for ionised hypocalcaemia. Women classified as having total crude hypocalcaemia had their odds of ionised hypocalcaemia increased by 5.88 folds compared with their counterparts with total normocalcaemia (p-value = 0.003). This significant association still brings to light the possibility of predicting ionised calcaemic states from total calcaemia. However, these equations are generally complex given that ionised calcium concentrations depend on several variables like the pH, total albumin, plasma proteins, and concentration of some ions in the blood. Even though total hypocalcaemia in pregnancy might exaggerate the burden, it is a probable predictor of the likelihood of ionised hypocalcaemia in these women. If women with total hypocalcaemia have their odds of ionised hypocalcaemia increased by about 6 folds, women diagnosed with total hypocalcaemia should be systematically checked for variations of the metabolically active fraction.

Apart from a monthly revenue below 100.000FCFA (179 USD), no sociodemographic factor had a significant association with total crude hypocalcaemia. Notwithstanding, a study carried out in Maiduguri teaching hospital (Nigeria) reported the level of education to be significantly associated with a reduced likelihood of total crude hypocalcaemia [4]. A similar trend was observed in this study but was not statistically significant. Our findings on the relationship between age and total hypocalcaemia also concord with results from Nigeria [4].

We found no studies evaluating the influence of monthly income on the likelihood of total hypocalcaemia in pregnancy. Our findings suggest that poorer women were less likely to have total crude hypocalcaemia than their rich counterparts. This relationship is complex given that from expectations, rich women should feed well and have money to buy calcium supplements than poor women. However, this effect could be associated with increased access of poor women to locally cultivated calcium-containing foodstuffs than the richer women who might have to buy everything from the market.

Notwithstanding, nutritional variables showed statistically significant associations with total crude hypocalcaemia. Women who took at least 3 meals a day had significantly reduced odds of total crude hypocalcaemia. This is logical as adequate nutritional intake is associated with an increased likelihood of meeting calcium demands. Moreover, taking desserts between meals reduced the odds of total crude hypocalcaemia by 0.73 folds. Desserts are usually dairy products or derivatives of dairy products and constitute any other food of fruits taken between meals. This additional intake of food/fruit can potentially lead to relatively more calcium intake a day. We found no studies that associated desserts consumption with low serum calcium in pregnancy.

Furthermore, consumption of tubers, plantains/bananas as opposed to cereals like rice and corn was associated with increased odds of total crude hypocalcaemia. Consumption of potassium-rich diets could indirectly induce pseudo-hypoparathyroidism and cause hypocalcaemia in pregnancy [41]. This could have been the case with consumption of meals with potassium-rich content like plantains and bananas [42]. According to a study carried out in Nkongsamba on the calcium content of some commonly consumed meals, cereal-containing meals were found to have relatively high calcium content compared with meals made from tubers, plantain and bananas [43]. However, in the same study, these meals contained relatively high amounts of phytates and moderate concentrations of oxalates which inhibit calcium absorption [43]. Studies targeting related nutritional matters need to be designed in this setting to evaluate the mineral content (particularly calcium) and relative concentrations of calcium absorption inhibitors (as well as the invitro calcium availability) of these tubers and cereals in both cooked and uncooked states.

Hence, this study highlights through these factors that adequate dietary intake is indispensable in meeting calcium needs in pregnancy. Pregnant women should be encouraged to take at least three meals a day, take desserts and consume more cereals to maximise the chances of averting hypocalcaemia. These recommendations align with WHO recommendations on nutrition in pregnancy [9].

From the viewpoint of ionised hypocalcaemia, women above 30 years, women with higher education, single women, women who initiated ANC before 4 months of pregnancy, housewives and women who did not take desserts between meals had their odds of ionised hypocalcaemia significantly increased. No research has evaluated the sociodemographic, obstetric and nutritional factors affecting ionised hypocalcaemia. Our findings suggest that even though ionised calcaemic states depend on physiological changes and total calcaemia, upstream factors (such as sociodemographic, obstetric and nutritional factors) may have a crucial role (direct or indirect) to play by influencing the physiological state of these women. More research is required to better elucidate the role played by these upstream factors on ionised calcaemic states, especially in pregnancy.

Although not statistically significant, calcium supplementation reduced the odds of having all forms of hypocalcaemia in pregnancy. The observed effect is clinically significant given that it is beneficial, acceptable, cost-effective and easy to implement despite its statistical insignificance [44]. This relationship between calcium supplementation and calcaemic states is at the heart of the WHO recommendations on calcium supplementation among women with insufficient dietary intake [9, 45, 46].

Taking calcium supplements together with other supplements was associated with reduced odds of total crude hypocalcaemia. As expected with ionised hypocalcaemia, the strength of association was weaker. This relationship could be explained from the viewpoint of adherence. Participants who took all their supplements at the same time are likely to have been more adherent to calcium supplementation, given that adherence to iron and folic acid supplements is higher among pregnant women [47-49]. Taking more than 1000mg of elemental calcium had a negative non-statistically significant association with total hypocalcaemia. Nevertheless, the association carried some tendency of statistical significance with a p-value of 0.061. This observation aligns with expectations and highlights the necessity of high dose supplementation suggested by the WHO. The tendency of statistical significance might be due to the sample size, but the strength of association is relatively considerable (OR = 0.75).

Surprisingly, women who took more than 1000mg of elemental calcium a day had their odds of ionised hypocalcaemia increased by 2.32 folds. This unexpected result contrasts with the findings on total crude hypocalcaemia given that, amongst others, ionised calcaemia depends on total calcaemia. This observation could be due to the small proportion of women with ionised hypocalcaemia and thus, warrants a more extensive exploration.

The results presented in this write-up should be exploited with full knowledge of the limits. We presented the prevalence for total crude, albumin-corrected and ionised hypocalcaemia in late pregnancy but failed to hormonally explore women with ionised hypocalcaemia to find full explanations. The physiological implications and consequences of hypocalcaemia were not addressed in our study. Some of the variables adopted as nutritional factors were not quantifiable and could not be applied directly. For example, fruit consumption was not evaluated in terms of type and number of pieces a day (which was very difficult to evaluate in our context). Furthermore, we did not make any difference between chronic hypocalcaemia and acute changes in calcaemic states. The cross-sectional design adopted in this research could allow for the detection of associations that are not permanent and hence, subject to variations. Moreover, no causation can be established using our design. However, the design remains the most practical design adapted for exploring such relationships.

Conclusion and recommendations

Ionised hypocalcaemia in pregnancy is a rare finding. Only 2.89% of all apparently healthy pregnant women have ionised hypocalcaemia in late pregnancy. However, the prevalence of total hypocalcaemia in late pregnancy among apparently healthy women in NHD is as high as 61.64% and 56.70% depending on whether total crude or total albumin-corrected calcium concentrations are considered, respectively. Monthly income below 100.000FCFA (USD 179), taking plantain/banana and tubers as the content of their most consumed meal (opposed to cereals), significantly increase the odds of total hypocalcaemia. Moreover, taking more than 2 meals a day and desserts between meals significantly reduce the odds of total hypocalcaemia in pregnancy.

On the other hand, factors like age above 30 years, having a higher education, being single, having initiated ANC before 4 months of pregnancy, being a housewife and not taking desserts between meals have significantly positive associations with ionised hypocalcaemia.

Total hypocalcaemia in pregnancy should not be neglected but should warrant measurement of the ionised fraction for better classification. Pregnant women with ionised hypocalcaemia should be thoroughly evaluated, and further hormonal exploration carried out. Even though calcium supplementation is a solution to meeting calcium needs in pregnancy, the fact that principally nutritional factors influence total hypocalcaemia suggests that increased consumption of locally available calcium-rich meals will help combat hypocalcaemia in pregnancy. Further studies should be carried out to understand the causes of ionised hypocalcaemia registered in pregnancy.

Data base of determinants and effects of low serum calcium in pregnancy.

(MDB)

Click here for additional data file.

7 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Frank T. Spradley

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Does the manuscript adhere to the experimental procedures and analyses described in the Registered Report Protocol?

If the manuscript reports any deviations from the planned experimental procedures and analyses, those must be reasonable and adequately justified.

Reviewer #1: Yes

**********

2. If the manuscript reports exploratory analyses or experimental procedures not outlined in the original Registered Report Protocol, are these reasonable, justified and methodologically sound?

A Registered Report may include valid exploratory analyses not previously outlined in the Registered Report Protocol, as long as they are described as such.

Reviewer #1: Yes

**********

3. Are the conclusions supported by the data and do they address the research question presented in the Registered Report Protocol?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. The conclusions must be drawn appropriately based on the research question(s) outlined in the Registered Report Protocol and on the data presented.

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General comments

This manuscript explores associations of low blood Ca in pregnant women and various sociodemographic, obstetric or nutritional factors. This cross-sectional study was conducted in 4 different health care settings in Cameroon.

Consider using the term low blood Ca in clinically healthy subject throughout the manuscript, instead of hypocalcemia. Please note that the title reads low blood Ca. Hypocalcemia connotates an abnormality with potentially negative clinical effects.

No reference has been cited to support the choice of low blood Ca/hypocalcemia thresholds.

There is a large disagreement in prevalence of low blood Ca, based on the type of the analytical method of choice. This issue deserves further attention. Please consider the following:

1. Visually evaluate your data by plotting iCa (consider the gold standard) vs total calcium, and iCa vs albumin-corrected total Ca.

2. Evaluate the correlation of iCa with tCa and albumin-corrected tCa.

3. Consider evaluating if there is a bias across lab methods, and if this bias is constant.

The criterion used to stratify the sociodemographic (i.e., age, monthly revenue, household occupants), obstetric (i.e., gestational age at first antenatal visit, number of visits, number of pregnancies, number of deliveries), and nutritional factors (i.e., taking fruit rare) should be clear. Describe the reasoning behind the choice of categories.

Authors do not provide enough background to understand the selected nutritional factors as explanatory variables within the context of Cameroon dietarian preferences. For example, it is unclear why authors decided on oranges, watermelons, and pineapples to classify the most consumed fruit. “Eating fruit rarely” it is not a quantifiable variable and may have different meaning across countries or ethnicities (i.e., number of pieces or servings per day).

Specific Comments

L 1. Acknowledge that the study was conducted in Cameroon.

L164. As indicated above, authors need to support the hypocalcemia thresholds with references (iCa, albumin-corrected tCa, and tCa).

L 164. Report the number of cows identified with low Ca serum when combining result from the 3 methods.

L 202. Please explain in materials how data was analyzed to calculate the prevalence of low calcium in blood.

L 217. Consider rephrasing as “Sociodemographic and obstetric factors associated with hypocalcemia are presented in Table 3.”

L 236. Change “Table 8: Nutritional factors affected” by “Table 4: Nutritional factors associated”.

L 250. The first discussion paragraph is written as results. Follow scientific writing guidelines for discussion, and delete it.

L 250. It is striking the difference in prevalence when considering different lab methods. That should be one of the first points of discussion.

L 260 to 262. Alkalogenic diets (rick in K) may induce pseudohypoparathyroidism. Please see http://doi.org/10.3618/jds.2013-7467

L 265. Replace “borderline significance” by “tendency to”. Define in material and methods tendency, for example p< 0.1

L 280. What was the prevalence with unadjusted Calcium levels?

L 296 to 299. Disagreement in Ca concentration across lab methods has been reported in patients with blood Ca metabolism disorders. However, the present study included healthy patients. Are there other plausible explanations? As indicated above, please explore the data.

L 306. Replace regnancies with pregnancies.

L 320. This section of the discussion focuses on VitD, but it was not measured in this study.

L 328. Do no start a new paragraph with Therefore.

L 401. Another limitation of the study is that the physiological implications of low blood calcium were not explored.

Tables

Remove the table’s superscripts indicating statistical significance (no needed). Explain in tables and manuscript if all the variables selected to be in multivariable model based on p-value, remained in the model or were removed after backward elimination.

The tables should stand on their own. Include the definition of hypocalcemia (lab method and threshold) used in the data analysis.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Jan 2022

General reviewers comments

We wish to appreciate the efforts and work done by the reviewer to strengthen this manuscript. The comments raised by the reviewer have gone a long way to significantly improve the quality of the manuscript. The response to the different reviewer’s comments are presented below.

This manuscript explores associations of low blood Ca in pregnant women and various sociodemographic, obstetric or nutritional factors. This cross-sectional study was conducted in 4 different health care settings in Cameroon.

Resp: Thanks for the observation.

Consider using the term low blood Ca in clinically healthy subject throughout the manuscript, instead of hypocalcemia. Please note that the title reads low blood Ca. Hypocalcemia connotates an abnormality with potentially negative clinical effects.

Resp: Thanks for your suggestion. Hypocalcaemia (generally used to mean “lower than normal blood calcium levels”) can either be an asymptomatic laboratory finding or a life-threatening metabolic disturbance (1)(2). We however agree on the precision “in clinically healthy or “apparently” healthy pregnant women” which has been used throughout the manuscript and made clear in the conclusion. This correction has been integrated in the whole manuscript where necessary.

No reference has been cited to support the choice of low blood Ca/hypocalcemia thresholds.

Resp: Thanks for the observations. The reference has been included. See page 10, first paragraph

There is a large disagreement in prevalence of low blood Ca, based on the type of the analytical method of choice. This issue deserves further attention. Please consider the following:

1. Visually evaluate your data by plotting iCa (consider the gold standard) vs total calcium, and iCa vs albumin-corrected total Ca.

2. Evaluate the correlation of iCa with tCa and albumin-corrected tCa.

3. Consider evaluating if there is a bias across lab methods, and if this bias is constant.

Resp: Thank you for your observations and suggestions. These were some original worries when the protocol was been designed. We intended to use ionised hypocalcaemia as a gold standard to evaluate the sensitivity of total calcaemia measurements in defining hypocalcaemia. We however found out that this was not scientifically correct given the ionised and total calcium are not exactly the same laboratory finding diagnosed by different methods. Multiple studies have agreed that even though total hypocalcaemia seems highly prevalent in pregnancy (laboratory finding), this constitutes some form of false or factitious hypocalcaemia associated with physiological changes in pregnancy that can be ignored (3). Ionised and total hypocalcaemia in pregnancy should therefore be studied as two entities as they might be associated with adverse outcomes of different severities.

However, the results of the requested additional analyses have been presented in the results section. See the two attached figures and page 14, paragraph 2 and page 15

Given that we are measuring different outcomes with different methods, we think that it will not make sense to assess bias across methods. Besides some studies suggest that different health conditions affect these different forms of calcium differently.

The criterion used to stratify the sociodemographic (i.e., age, monthly revenue, household occupants), obstetric (i.e., gestational age at first antenatal visit, number of visits, number of pregnancies, number of deliveries), and nutritional factors (i.e., taking fruit rare) should be clear. Describe the reasoning behind the choice of categories

Resp: Thank you for your question. Given that we did not find any similar studies carried out with categorization. Most of the categorization was done from previous studies carried out in pregnancy and particular WHO health indicators. Some of the variables were categorized following standard divisions like in a union/not in a union.

Concerning fruit consumption, people who consumed fruits rarely were expected to have hypocalcaemia compared with their counterparts who do more often (4).

See paragraph 2, page 10.

Authors do not provide enough background to understand the selected nutritional factors as explanatory variables within the context of Cameroon dietarian preferences. For example, it is unclear why authors decided on oranges, watermelons, and pineapples to classify the most consumed fruit. “Eating fruit rarely” it is not a quantifiable variable and may have different meaning across countries or ethnicities (i.e., number of pieces or servings per day).

Resp: Thanks for your suggestion. Studies around nutritional habits and practices in Cameroon are still very few. Studies carried out in pregnant women in Cameroon are ever rarer. We did not have any published studies on dietarian practices.

Concerning fruits, Consumption of vegetables and fruits has been reported to better calcium availability in blood and bone mineral density. We therefore believed that women who consumed fruits rarely were likely to have lower calcaemia compared to their counterparts who consumed fruits more often (4).

We recognize the fact that this variable is not quantifiable and will include this in the limits of the study. In our context, it is really difficult to evaluate this in pieces or servings per day). See limits section, page 30, paragraph 2.

As concerns the most consumed fruits, only pineapples, oranges, watermelons and ripe banana were specified in the questionnaire and all other fruits included into others. When analysed without categories, no particular fruit had a statistically significant association with hypocalcaemia. As stated in the background, the most consumed fruits were watermelon, oranges and pineapples. We therefore wanted to know if consumption of the most frequently consumed fruits in Cameroon was associated with hypocalcaemia. Some information on the nutritional behaviors has been added to the background. See page 5, paragraph 3.

Specific Reviewers comments Response to comments

L 1. Acknowledge that the study was conducted in Cameroon.

Resp: Thanks for the recommendation. Cameroon has been added to the title. See line 1

L164. As indicated above, authors need to support the hypocalcemia thresholds with references (iCa, albumin-corrected tCa, and tCa).

Resp: Thanks for your recommendations, the references have been added. Already addressed above

L 164. Report the number of cows identified with low Ca serum when combining result from the 3 methods.

Resp: Thank you for the recommendation. The number of women meeting the criteria has been added. Only 28 participants met this criteria. See results section, page 14, paragraph 1.

L 202. Please explain in materials how data was analyzed to calculate the prevalence of low calcium in blood.

Resp: Thank you for the suggestion. The cut-offs considered were stated clearly in the data analysis section. All women who had ionised, total crude and albumin-corrected total calcium levels less than or equal to the stated cut-offs were classified as hypocalcaemic (low blood calcium). This statement has been included in the data analysis section. See page 10, paragraph 1.

L 217. Consider rephrasing as “Sociodemographic and obstetric factors associated with hypocalcemia are presented in Table 3.”

Resp: Thank you for your correction. Adopted in the text. See page 16, paragraph 1.

L 236. Change “Table 8: Nutritional factors affected” by “Table 4: Nutritional factors associated”.

Resp: Thank you for the correction. Corrected accordingly. See table 4, page 19.

L 250. The first discussion paragraph is written as results. Follow scientific writing guidelines for discussion, and delete it.

Resp: Thank you for your recommendation. It has been deleted and reformulated. See paragraph 1, page 21.

L 250. It is striking the difference in prevalence when considering different lab methods. That should be one of the first points of discussion.

Resp: Thanks for the suggestion. The discussion has been shifted up to start by discussing the discrepancy. The laboratory methods can really not be discussed here given that these are recommended and gold standard procedures especially for ISE potentiometry. The methods adopted (AAS and ISEP) measure two different targets (ionised calcium and total calcium). Ideally ionised (physiological) hypocalcaemia should be the true target and not the total hypocalcaemia, even though having total hypocalcaemia increased the chances of having ionised hypocalcaemia (stated in the discussion). See page 27, paragraph 2. As stated above additional analysis has been presented.

L 260 to 262. Alkalogenic diets (rick in K) may induce pseudohypoparathyroidism. Please see http://doi.org/10.3618/jds.2013-7467

Resp: Thanks for your suggestion. This has been included in the discussion. See page page 28, last paragraph.

L 265. Replace “borderline significance” by “tendency to”. Define in material and methods tendency, for example p< 0.1

Resp: Thanks for the suggestion. We prefer to state clearly that the association was not statistically significant. And include the tendency in the next sentence. See page 30, paragraph 1.

L 280. What was the prevalence with unadjusted Calcium levels?

Resp: Thank you for the question. The prevalence with unadjusted calcium levels is the prevalence of totalcrude hypocalcaemia. That is, when the total calcium levels are not adjusted or corrected for albumin changes. This has been reported in the results and first paragragh of the discussion. The word “unadjusted” has been changed and made more explicit. See page 24, paragraph 2.

L 296 to 299. Disagreement in Ca concentration across lab methods has been reported in patients with blood Ca metabolism disorders. However, the present study included healthy patients. Are there other plausible explanations? As indicated above, please explore the data.

Resp: Thanks for your concern. As already explained above, only two diagnostic methods were used and these two diagnostic methods did not measure the same calcium fractions in blood. One measured and classified calcium levels based on the physiologically active fraction which is the absolute hypocalcaemia, while the other method measured the whole calcium in circulation (which can be considered a relative hypocalcaemia). See page 21, paragraph 2, sentence 3 and 4, of discussion

L 306. Replace regnancies with pregnancies.

Resp: Thank you for the correction. This has been corrected. See page 22, first paragraph.

L 320. This section of the discussion focuses on VitD, but it was not measured in this study.L 328. Do no start a new paragraph with Therefore.

Resp: Thanks for the observation. Vit D deficiency and other hormonal disorders were not measured in this study (stated as a limit). The discussion on these was just to identify based on relative prevalence from other studies, the plausible causes of the hypocalcaemia (particularly the ionised hypocalcaemia) in pregnancy. The “therefore” has been taken off the sentence. See page 23, paragraph 2

L 401. Another limitation of the study is that the physiological implications of low blood calcium were not explored.

Resp: Thanks for the suggestion. This has been included as a limit. See page 30, last paragragh.

Tables

Remove the table’s superscripts indicating statistical significance (no needed). Explain in tables and manuscript if all the variables selected to be in multivariable model based on p-value, remained in the model or were removed after backward elimination.

The tables should stand on their own. Include the definition of hypocalcemia (lab method and threshold) used in the data analysis.

Resp: Thanks for the recommendations. The tables have been edited accordingly. See tables 3, 4 and 5.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

14 Feb 2022

19 Feb 2022

RESPONSE TO REVIEWERS

We wish to thank the editor and reviewer for the time taken to improve on this manuscript. We have now taken time to address all the editor and reviewer comments. This is presented below.

Editor’s comments

Comment: Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Response: Thanks for your advice. The references have been checked and some changes made. Ref number 8 and 11 have been changed, and one of the references which was a repetition of reference number 3 taken off. See reference list.

Reviewer’s comments

Comment: Thanks to the authors for revising the manuscript. It has greatly improved. I have additional comments that may help improve the quality of this work.

Response: Thanks for your observation. The efforts put in by the reviewer to improve this work are highly appreciated on our side. We must agree that the review process has highly improved on the quality of this manuscript.

Comment: Consider that Plos One does not copyedit manuscripts. The language used is correct but not very elegant. I will strongly recommend a professional editor to polish the manuscript prior to publication.

Response: Thanks for the observation. We had our manuscript screened by a native English speaker and hope the level of English is up to standard now.

Major comment: Why the study focuses on hypocalcemia based on total Ca (tCa) instead of ionized Ca (iCa)? Should be more appropriate to present associations with both iCa and tCa? The number of subjects with low iCa is limited but enough to explore possible associations.

Response: We highly appreciate the reviewers question which is pertinent. We belief that leaving this out would have taken a big chunk off this work. As recommended, we have added analysis for ionised calcaemia and changes made on the whole manuscript to reflect the analyses. This has affected the objective statement, the results, discussion and conclusion. See abstract (page 2 and 3), last paragraph of introduction (page 6), results section (page 21, 22 and 24, including table 5, 6 and 7), discussion section (page 32, second paragraph and page 33, last paragraph), conclusion section, first paragraph, page 35).

Minor comment: Revise the manuscript and use concentration instead level. Be consistent.

Response: Thanks for your correction. The whole manuscript has been revised as recommended.

Specific comments:

Comment: L 34. Please consider rephrasing: Blood samples were collected and analyzed for serum ionised calcium concentrations and pH (ion-selective potentiometry), and for total calcium and albumin concentration (atomic absorption spectrophotometry). Sociodemographic, obstetric and nutritional data were collected through interviewer-administered questionnaire.

Response: Thanks for taking time to correct this. We have adopted it as recommended. See Abstract, page 2, line 35-39.

Comment: L57 – 66. Indicate the type of analysis (iCa or tCa) used to evaluate hypocalcemia in studies 1 to 5.

Response: Thanks for your suggestion. This has been added to the first paragraph of the introduction.

Comment: L 167. Explain in this section the software used to do the statistical analysis, and methods for regression and logistic regression.

Response: Thanks for your suggestion: All these are included in the data analysis section. See page 9, 10 and 11.

Comment: L 167. Explain in the data analysis section what measurement and definition of hypocalcemia was used to build the multivariable models.

Response: Thanks for your suggestion. This has been included in the data analysis section. See page 10, paragraph 2.

Comment: L 248 to 251. Is this information a repetition of what presented in Fig 1 and 2? If so, delete.

Response: Thanks for your observation. This is not a repetition but additional information not accessible on the figures. We think the information should remain.

Comment: L 259. Consider using the word odds when reporting results. For example “Participants who earned less than 100.000FCFA (179 USD) in a month had 0.73 lower odds to have total crude hypocalcaemia”. Revise entire manuscript. This is important because this study reports odds no likelihood or risks.

Response: Thank you very much for pulling our attention to that. We have revised the whole manuscript accordingly.

Comment: L 308 discusses de physiological implications of iCa but the study evaluates associations with tCa.

Response: Thanks for your observation. We have now included associations with iCa in the manuscript.

Comment: L 293 to 295. This is results. It does not belong in the discussion.

Thanks very much, it has been taken up to the results section. See result section, page 16, line 286-288.

Comment: Table 3 and 4 footnotes, modified as:

Response: Thanks for your suggestion. This has been modified.

Submitted filename: RESPONSE TO REVIEWERS.docx

Click here for additional data file.

26 Apr 2022

27 Apr 2022

RESPONSE TO REVIEWERS

We appreciate the immense effort of the editor and the reviewer towards strengthening this work. We have addressed the comments raised by the editor and reviewer below.

Editor’s comment: Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Response: Thanks for the recommendation. The reference list had been reviewed accordingly during the last revision and the changes were specified in the rebuttal letter.

Reviewer’s comments

Comment: Thank you for your effort. Please see below some minor comments.

Response: Thank you again for additional comments. We are grateful for you pertinent contributions

Comment: L 254 to 257 Please note that your sample size is very large and results in significant differences, but the correlation coefficient is poor.

Response: Thank you for the observation. We have taken this precision off given that it is not part of the objectives of this manuscript. However, we feel that not all statistical observations have direct clinical or biological implications. We already stated in the manuscript that the correlation was weak (poor). See line 254-257.

Comment: L 262 to 263 Your sample size is very large and results in significant differences, but it may have limited biological implications.

Response: Thank you for the comment; given that this is not part of any of the objectives of the manuscript, the specification has been removed. See results section, line 262-267.

Comment: L 332. Replace “Piece of work” with “The main goal of our study was to estimate the prevalence of …”

Response: Thank you for your suggestion. It has been adopted. See discussion section, line 332.

Comment: All the figures and tables should stand on their own without the need to gather additional information from the main body of the manuscript. Add the number of subjects in the total and ionized hypocalcemia. Please revise the space around equal signs in the tables and figures.

Response: Thank you for your comments; we have now made the requested additions to the tables and manuscript.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

5 May 2022

Ionised and total hypocalcaemia in pregnancy: an analysis of prevalence and risk factors in a resource-limited setting, Cameroon

PONE-D-21-31669R3

Dear Dr. Ajong,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Frank T. Spradley

Academic Editor

PLOS ONE

10 May 2022

PONE-D-21-31669R3

Ionised and total hypocalcaemia in pregnancy: an analysis of prevalence and risk factors in a resource-limited setting, Cameroon

Dear Dr. Ajong:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Frank T. Spradley

Academic Editor

PLOS ONE