Menstrual characteristics of sickle cell disease patients seen at a tertiary institution in North Western Nigeria.

Introduction: The burden of sickle cell disease (SCD) is high in Sub-Saharan Africa, including Nigeria, and with improved care and survival, reproductive health issues, including menstruation, have become increasingly important and may impact on long-term health and social wellbeing.
Objectives: This study was carried out to characterize the menstrual characteristics of SCD patients. Methodology: Using a cross-sectional study design, a semi-structured questionnaire was administered and information collected on bio-demographic data, medical history of SCD, reproductive, and menstrual history.
Results: One hundred and sixty female patients participated in the study. The mean age was 24.9 years (standard deviation [SD] ± 8.8). The mean age at menarche was 15.6 years (SD ± 2.7) for those menstruating. Most respondents had regular menstrual cycles 120 (82%); normal menstrual cycle lengths 120 (81%) significantly associated with respondent's age (P < 0.05); normal duration of flow 140 (97%); light menstrual flow 104 (71%) significantly associated with body mass index BMI (P < 0.05). Overall, however, 114 (78%) had abnormal menses and 32 (22%) had normal menses. Eighty-one respondents (56%) experienced menstrual pain, which was severe in only 28 cases (35%). The presence of menstrual pain did not significantly affect the annual frequency of crisis (P > 0.05).
Conclusion: The mean age at menarche was high at 15.6 years. Most respondents had light menstrual flow and overall abnormal menstrual patterns. Menstrual pain was common but was not significantly associated with the frequency of crisis.

INTRODUCTION

Sickle cell disease (SCD) is a group of autosomal recessive disorders, with a variety of phenotypes caused by mutations in the β hemoglobin gene, resulting in the polymerization of hemoglobin in hypoxic states, sickling of red blood cells, vaso-occlusion and hemolysis which consequently cause variable complications including end-organ damage and severe morbidity or mortality.[12]

The burden of SCD is high in Sub-Saharan Africa, accounting for over 75% of the estimated 300,000 children born with SCD annually worldwide.[3] In Nigeria, the prevalence of SCD is about 2%–3%, with an estimated 150,000 new births annually.[456789]

Advances in medicine have led to improved care and survival of women with SCD; hence, reproductive health issues, including menstruation, have become increasingly important.[10] Menstruation may also impact on social well-being, and be an important marker of current and potential health problems.[11] Menstrual cycle parameters can serve as noninvasive clinical markers of reproductive function be used to monitor environmental effects on reproduction and predict long-term health outcomes, including breast cancer and cardiovascular disease risk factors.[1213141516]

As with all women, menstrual characteristics may be varied in women with SCD. However perhaps, the disease may directly influence the menstrual characteristics and vice versa. Menstruation is also associated with pain, which may affect SCD-related pain rate, though studies are inconsistent.[101718]

This study was done to characterize the menstrual characteristics of SCD patients. Few studies have been done in this area, especially in this environment and may be useful to improve patient counseling and management.

METHODOLOGY

Description of the study area

The study setting was the Barau Dikko Teaching Hospital (BDTH), of the Kaduna State University, North-Western Nigeria. The hospital serves the people of Kaduna metropolis and its environs. The adult SCD clinic runs under the department of Haematology each week, with an average of 25–30 patients seen, bearing in mind an equal sex incidence in SCD.

Study design

The study was a cross-sectional descriptive study involving the administration of a semi-structured questionnaire.

Study population

Participants for the study were drawn from female patients diagnosed with SCD attending the weekly sickle cell clinic either newly, or for follow-up care.

Inclusion criteria

Diagnosed SCD confirmed by either Haemoglobin electrophoresis in alkaline medium or high-performance liquid chromatography, female, aged 15 years or above, consented.

Exclusion criteria

Patients on hormonal medications were excluded.

Sample size calculation

A convenient sample of all eligible female SCD patients consenting to participate in the study was taken.

Sampling technique

Purposive sampling was used.

Data collection

A pretested semi-structured questionnaire was administered, and information collected included bio-demographic data, medical history of SCD, reproductive, and menstrual history.

Definition of terms

For the purpose of this study, normal menstrual parameters assessed were defined as follows;[1920]

Menarche

Menarche defined as age at which first menses commenced. Normal age at menarche was between the ages of 10–16 years.

Cycle regularity

Cycle regularity defined as cycle to cycle variation over 12 months. This was considered regular if it was between 2 and 20 days.

Cycle length

Cycle length defined as the number of days from the 1st day of one menstrual period to the 1st day of the next menstrual period, this was considered normal if it was between 21 and 35 days.

Cycle duration

Cycle duration defined as the number of days a woman menstruated for within one cycle. This was considered normal if it was for 2–8 days.

Quantity of menstrual bleeding

Quantity of menstrual bleeding defined as volume of blood loss during menstruation. This was estimated indirectly using the number of sanitary pads/tampons or napkins used and was considered normal if 3–6 pads were used per day.

Normal menstruation

If age at menarche, cycle regularity, cycle length, cycle duration and quantity of bleeding were within the defined normal parameters.

Abnormal menstruation

If one or more of the defined parameters (age at menarche, cycle regularity, cycle length, cycle duration, and quantity of bleeding) were abnormal.

Respondents also scored the severity of their menstrual pain (dysmenorrhea) on a scale of 1–10 and this was reclassified as mild (score of 1–3), moderate (score of 4–6), and severe (score of 7–10).

Menopause (natural menopause)

The permanent cessation of menstruation for up to 12 months without any obvious physiologic or pathologic cause.

Premature menopause

Menopause before ≤40 years.

Body mass index (BMI) was used as a measure of adiposity, calculated from participants’ height and weight (kg/m2) and interpreted based on standard classifications (National Institutes of Health 1998).

Data analysis

Data were analyzed with IBM Statistical Package for the Social Sciences (IBM SPSS Statistics; Armonk, NY, USA) software version 22. Data were summarized using cross tab and frequency tables. Chi-square or Likelihood Ratio test as appropriate was used to check for a statistically significant association between the dependent and independent variables. Binary logistic regression was used to determine which variables were associated with normal and abnormal menstruation with the following covariates: BMI (normal and abnormal), age (≥35 or <35), occupation (government employed or unemployed/self-employed), ethnicity (Hausa and non-Hausa), religion (Islam or Christianity), education (primary or less and secondary or more), and parity (nulliparous or parous). A value of P < 0.05 was considered statistically significant.

Ethics

Ethical approval for the study was gotten from the BDTH Health Research Ethics committee and signed informed consent from participants. The nature of the study posed very little risk to participants whose details were kept confidential.

RESULTS

One hundred and sixty female patients participated in the study from November 2018 to August 2019. All respondents had a hemoglobin phenotype of HbSS. Characteristics of the study population are shown in Table 1. The mean age was 24.9 years (standard deviation [SD] ± 8.8). Majority of respondents had at least secondary level education or higher (147, 91.9%), were mostly students 90 (56.2%), single 125 (78.1%) and nulliparous 130 (81.3%) [Table 1]. Their mean weight was 49.7 kg, and the mean BMI was 19.7 (SD ± 9.3 kg and 3.4, respectively). Menstrual characteristics are shown in Table 2; only 146 respondents (91%) had attained menarche at the time of the study.

Table 1

General characteristics of study population

Characteristic	Frequency (n=160; 100%), n (%)
Age
≤9	55 (34.3)
20-29	62 (38.7)
30-39	31 (9.4)
40-49	10 (6.3)
≥50	2 (1.3)
Education
None	1 (0.6)
Quaranic only	5 (3.1)
Primary	7 (4.4)
Secondary	82 (51.3)
Tertiary	65 (40.6)
Occupation
Business woman	14 (8.8)
Civil servant	21 (13.1)
Full time house wife	17 (10.6)
Petty trader	18 (11.3)
Student	90 (56.2)
Religion
Christianity	50 (31.2)
Islam	110 (68.8)
Ethnicity
Hausa	85 (53.1)
Yoruba	9 (5.6)
Igbo	0 (0)
Other	66 (41.3)
Marital status
Married	30 (18.8)
Single	125 (78.1)
Widowed	1 (0.6)
Divorced/separated	4 (2.5)
Parity
0	130 (81.3)
1-2	25 (15.6)
3-4	4 (2.5)
≥5	1 (0.6)
BMI
<18.5	62 (38.8)
18.5-24.9	89 (55.6)
25-29.9	6 (3.8)
≥30	3 (1.8)

BMI: Body mass index

Table 2

Menstrual characteristics

Characteristic	Frequency (n=160; 100%), n (%)
Age at menarche (n=160)
Never menstruated	14 (8.8)
<10 years	0
10-16 years	102 (63.7)
>16 years	44 (27.5)
Cycle regularity (n=146)
Regular	120 (82.2)
Irregular	24 (16.4)
Don’t know	2 (1.4)
Cycle length (days) (n=146)
<21	8 (5.4)
21-35	118 (80.8)
>35	20 (13.8)
Duration of menstrual flow (days) (n=146)
<2	5 (3.4)
2-8	141 (96.6)
>8	0
Average number of sanitary pads used per day (n=146)
<3 (light)	104 (71.2)
3-6 (normal)	42 (28.8)
>6 (heavy)	0
Menstrual pain (n=146)
No	65 (44.5)
Yes	81 (55.5)
Severity of menstrual pain (n=81)
Mild	29 (35.8)
Moderate	24 (29.6)
Severe	28 (34.6)
Last menses ≥12 months
Yes	8
No	138

Fourteen respondents (8.8%), whose ages ranged from 15 to 35 years, had not attained menarche. One hundred and two respondents (64%) attained menarche between the ages of 10–16 years. The mean age at menarche was 15.6 years (SD ± 2.7) for those menstruating. BMI and age were significantly associated with the age at menarche (P < 0.05) [Tables 2 and 3]. Eight respondents whose age ranged from 20 to 58 years had currently not seen their menstruation for up to 12 months and can be considered to be postmenopausal, prematurely in five of them.

Table 3

Factor that maybe associated with normal or abnormal menstrual characteristics

Characteristic (n=160)	Abnormal age at menarche (n=58) (row %)	Normal age at menarche (n=102) (row %)	Statistical test
BMI
Underweight	31 (50)	31 (50)	χ2=8.425, df=2, P=0.015
Normal	24 (27)	65 (73)
Overweight	3 (33.3)	6 (66.7)
Age
<19	18 (37.7)	37 (67.3)	Likelihood ratio=10.397, df=4, P=0.034
20-29	16 (25.8)	46 (74.2)
30-39	18 (58.1)	13 (41.9)
40-49	5 (50)	5 (50)
>50	1 (50)	1 (50)
Education
Primary or less	8 (61.5)	5 (38.5)	Likelihood ratio=3.736, df=1, P=0.053
Secondary or more	50 (34)	97 (66)
Religion
Christianity	16 (32)	34 (68)	χ2=0.568, df=1, P=0.451
Islam	42 (38.2)	68 (61.8)
Occupation
Business woman	7 (50)	7 (50)	χ2=6.265, df=4, P=0.180
Civil servant	5 (23.8)	16 (76.2)
Housewife	7 (41.2)	10 (58.8)
Petty trader	10 (55.6)	8 (44.4)
Student	29 (32.2)	61 (67.8)
Ethnicity
Hausa	33 (38.8)	52 (61.2)	χ2=1.006, df=2, P=0.587
Yoruba	4 (44.4)	5 (55.6)
Others	21 (31.8)	45 (68.2)
Parity
Nulliparous	46 (35.4)	84 (64.6)	χ2=0.225, df=1, P=0.635
Parous	12 (40)	18 (60)
Annual number of crisis
<3	27 (34.2)	52 (65.8)	χ2=0.290, df=1, P=0.590
≥3	31 (38.3)	50 (61.7)
Characteristic (n=144)	Abnormal cycle regularity (n=24) (row%)	Normal cycle regularity (n=120) (row %)	Statistical test
BMI
Underweight	10 (19.6)	41 (80.4)	χ2=0.379, df=2, P=0.828
Normal	15 (17.9)	69 (82.1)
Overweight	1 (11.1)	8 (88.9)
Age
<19	11 (27.5)	29 (72.5)	Likelihood ratio=10.397, df=4, P=0.034
20-29	6 (9.8)	55 (90.2)
30-39	6 (19.4)	25 (80.6)
40-49	1 (10)	9 (90)
>50	0 (0)	2 (100)
Education
Primary or less	1 (10.0)	9 (90.0)	Likelihood ratio=0.387, df=1, P=0. 558
Secondary or more	23 (17.2)	111 (82.8)
Religion
Christianity	9 (20.0)	36 (80.0)	χ2=0.524, df=1, P=0.469
Islam	15 (15.2)	84 (84.8)
Occupation
Business woman	0	14 (100)	Likelihood ratio=9.382, df=4, P=0.052
Civil servant	4 (19.0)	17 (81.0)
Housewife	1 (5.9)	16 (94.1)
Petty trader	2 (11.1)	16 (88.9)
Student	17 (23.0)	57 (77.0)
Ethnicity
Hausa	14 (18.4)	62 (81.6)	χ2=0.388, df=2, P=0.823
Yoruba	1 (12.5)	7 (87.5)
Others	9 (15.0)	51 (85.0)
Parity
Nulliparous	20 (17.5)	94 (82.5)	χ2=0.303, df=1, P=0.582
Parous	4 (13.3)	26 (86.7)
Annual number of crisis
<3	10 (13.9)	62 (86.1)	χ2=0.800, df=1, P=0.371
≥3	14 (19.4)	58 (80.6)
Characteristic (n=146)	Abnormal cycle length (n=28) (row %)	Normal cycle length (n=118) (row%)	Statistical test
BMI
Underweight	11 (21.2)	41 (78.8)	χ2=0.516, df=2, P=0.773
Normal	16 (18.8)	69 (81.2)
Overweight	1 (11.1)	8 (88.9)
Age
<19	16 (38.1)	26 (61.9)	Likelihood ratio=13.837, df=4, P=0.008
20-29	7 (11.5)	54 (88.5)
30-39	4 (12.9)	27 (87.1)
40-49	1 (10)	9 (90)
>50	0 (0)	2 (100)
Education
Primary or less	2 (20.0)	8 (80.0)	Likelihood ratio=0.005, df=1, P=0.945
Secondary or more	26 (19.1)	110 (80.9)
Religion
Christianity	9 (19.6)	37 (80.4)	χ2=0.006, df=1, P=0.936
Islam	19 (19.0)	81 (81.0)
Occupation
Business woman	0	14 (100)	Likelihood ratio=11.389, df=4, P=0.067
Civil servant	2 (9.5)	19 (90.5)
Housewife	2 (11.8)	15 (88.2)
Petty trader	3 (16.7)	15 (83.3)
Student	21 (27.6)	55 (72.4)
Ethnicity
Hausa	16 (20.8)	61 (79.2)	χ2=0.607, df=2, P=0.738
Yoruba	2 (25.0)	6 (75.0)
Others	10 (16.4)	51 (83.8)
Parity
Nulliparous	25 (21.6)	91 (78.4)	χ2=2.052, df=1, P=0.152
Parous	3 (10.0)	27 (90)
Annual number of crisis
<3	12 (16.7)	60 (83.3)	χ2=0.578, df=1, P=0.447
≥3	16 (21.6)	58 (78.4)
Characteristic (n=146)	Abnormal duration of flow (n=5) (row %)	Normal duration of flow (n=141) (row %)	Statistical test
BMI
Underweight	2 (3.8)	50 (96.2)	χ2=1.931, df=2, P=0.381
Normal	2 (2.4)	83 (97.6)
Overweight	1 (11.1)	8 (88.9)
Age
<19	1 (2.4)	41 (97.6)	χ2=9.108, df=4, P=0.058
20-29	1 (1.6)	60 (98.4)
30-39	1 (3.2)	30 (96.8)
40-49	2 (20)	8 (80)
>50	0 (0)	2 (100)
Education
Primary or less	0 (0.0)	1 (100.0)	Likelihood ratio=0.722, df=1, P=0.395
Secondary or more	5 (3.7)	131 (96.3)
Religion
Christianity	3 (6.5)	43 (93.5)	Likelihood ratio=1.781, df=1, P=0.182
Islam	2 (2.0)	98 (98.0)
Occupation
Business woman	0	14 (100)	Likelihood ratio=4.438, df=4, P=0.350
Civil servant	2 (9.5)	19 (90.5)
Housewife	1 (5.9)	16 (94.1)
Petty trader	1 (5.6)	17 (94.4)
Student	1 (1.3)	75 (98.7)
Ethnicity
Hausa	1 (1.3)	76 (98.7)	Likelihood ratio=2.942, df=2, P=0.230
Yoruba	1 (12.5)	7 (87.5)
Others	10 (16.4)	51 (83.8)
Parity
Nulliparous	2 (1.7)	114 (98.3)	Likelihood ratio=3.856, df=1, P=0.050
Parous	3 (10.0)	27 (90)
Annual number of crisis
<3	4 (5.6)	68 (94.4)	Likelihood ratio=2.077, df=1, P=0.149
≥3	1 (1.4)	73 (98.6)
Characteristic (n=146)	Abnormal quantity of flow (n=104) (row%)	Normal quantity of flow (n=42) (row%)	Statistical test
BMI
Underweight	38 (73.1)	14 (26.9)	χ2=6.740, df=2, P=0.034
Normal	63 (74.1)	22 (25.9)
Overweight	3 (33.3)	6 (66.7)
Age
<19	34 (81.0)	8 (19.0)	Likelihood ratio=4.480, df=4, P=0.345
20-29	42 (68.9)	19 (31.1)
30-39	19 (61.3)	12 (38.7)
40-49	8 (90)	2 (20)
>50	1 (50)	1 (50)
Education
Primary or less	6 (60.0)	4 (40.0)	Likelihood ratio=0.623, df=1, P=0.430
Secondary or more	98 (72.1)	38 (27.9)
Religion
Christianity	34 (73.9)	12 (26.1)	χ2=0.235, df=1, P=0.628
Islam	70 (70.0)	30 (30.0)
Occupation
Business woman	8 (57.1)	6 (42.9)	χ2=5.359, df=4, P=0.252
Civil servant	13 (61.9)	8 (38.1)
Housewife	12 (70.6)	5 (29.4)
Petty trader	11 (61.1)	7 (38.9)
Student	60 (78.9)	16 (21.1)
Ethnicity
Hausa	53 (68.8)	24 (31.2)	χ2=1.274, df=2, P=0.529
Yoruba	7 (87.5)	1 (12.5)
Others	44 (72.1)	17 (27.9)
Parity
Nulliparous	84 (72.4)	32 (27.6)	χ2=0.384, df=1, P=0.535
Parous	20 (66.7)	10 (33.3)
Annual number of crisis
<3	49 (68.1)	23 (31.9)	χ2=0.700, df=1, P=0.403
≥3	55 (74.3)	19 (25.7)

n: Number of respondents. BMI: Body mass index

One hundred and twenty respondents (82%) had regular menstrual cycles. None of the factors (age, occupation, education, religion, ethnicity, BMI, parity, or annul crisis frequency) was significantly associated with menstrual cycle regularity (P > 0.05) [Tables 2 and 3].

One hundred and eighteen respondents (81%) also had normal menstrual cycle lengths. Cycle length was significantly associated with respondent's age (P < 0.05) [Tables 2 and 3].

One hundred and forty-one respondents (97%) had a normal duration of flow. None of the factors (age, occupation, education, religion, ethnicity, BMI, parity, or annual crisis frequency) were significantly associated with duration of menstrual flow (P > 0.05) [Tables 2 and 3].

One hundred and four respondents (71%) had light menstrual flow. BMI was significantly associated with the quantity of menstrual flow (P < 0.05) [Tables 2 and 3].

When all the parameters of the menstrual cycle were considered in a combined manner (age at menarche, cycle regularity, cycle length, cycle duration, and quantity of bleeding), 114 (78%) had abnormal menses, and 32 (22%) had normal menses. In the logistic regression model, the covariates did not have statistically significant associations with normal or abnormal menstrual patterns [Table 4].

Table 4

Binary logistic regression of variables associated with normal and abnormal menstruation

Variable	OR	95% CI	P
BMI	0.503	0.23-1.09	0.080
Age	0.443	0.14-1.45	0.179
Education	0.694	0.17-2.80	0.608
Occupation	2.767	0.70-10.87	0.145
Religion	1.111	0.38-3.26	0.848
Ethnicity	0.943	0.34-2.60	0.910
Parity	1.551	0.58-4.13	0.379

OR: odds ratio, CI: Confidence interval

Eighty-one respondents (56%) experienced menstrual pain. Among those experiencing menstrual pain, only twenty-eight respondents (35%) considered it to be severe. Forty-six respondents used some form of medication to relieve their menstrual pain; Buscopan-4, Ibuprofen-15, Paracetamol-20, Pentazocin-1, Tramadol-1, unspecified-5. Other measures included sleep-6, use of warmth-4, hot water bottles-1, ingestion of ginger, garlic, and Lipton tea-2. The presence of menstrual pain did not significantly affect the annual frequency of crisis [Likelihood ratio 0.099, df-1, P = 0.753, Table 5].

Table 5

Cross tabulation comparing menstrual pain and frequency of crisis

Frequency of crisisannually	Menstrual pain (n=46)		Statistics
	No (%)	Yes (%)
1-2 times/year	33 (45.8)	39 (54.2)	Likelihood ratio - 0.099 df - 1, P - 0.753
≥3 times/year	32 (43.2)	42 (56.8)

DISCUSSION

This is one of the few studies that have examined the menstrual characteristics of patients with SCD, especially in this environment where prevalence is high. Our study population was fairly educated, and mostly students probably because this was a hospital based study, those that are educated are probably more likely to seek specialised hospital care to manage their SCD and its complications, and the hospital lies in close proximity to tertiary educational institutions in the state. Respondents were also mainly Hausa, which is the prevalent ethnic group at the study location. Curiously Igbos, one of the major ethic groups in Nigeria, were not among the respondents. The reason for this is unclear, but may perhaps have to do with higher rates of premarital screening. One study in southern suggested that a very low percentage of the couples (2.8%) had a 1 in 4 risk of their offspring inheriting SCD, and majority (2/3rd of the study population) would call-off their wedding if there was a risk of their offspring inheriting SCD.[21]

Primary amenorrhoea is said to occur in a girl by the age of 14–16 years (depending on the presence of other pubertal changes) that has not attained menarche and evaluation is required. About 9% of our study population had primary amenorrhoea. The mean age at menarche in our study was 15.6 years and this is on the high end of normal, though we did not consider other signs of puberty. In older studies, Samuels-Reid and Scott (1985)[22] found some delay in the onset of menarche among girls with SCD compared to control subjects (13.7 years vs. 13.1 years), which was similar to other reports.[2324252627] Some Zago et al.[28] showed that 56.2% of Brazilian girls aged 14 years with SCD were yet to attain menarche; the average age of menarche was 15 years which was higher than 13 years in those without SCD. This was similar to another Brazilian study that found an average age of menarche in patients with SCD to be 15 years and 8 months.[29] A delay in physical or sexual maturation may not be altogether a surprising finding in SCD and may accompany chronic anaemia and lower hormone production, especially growth hormone in the HbSS patient in childhood.[30]

In our study, BMI and age significantly affected the age at menarche. BMI and other anthropometric variables have been found to be lower in children with SCD when compared with children without SCD.[31] It seems a critical percentage of the body weight must be fat for the occurrence of menarche, to prevent weight-related amenorrhoea and maintain ovulatory cycles,[32] among all groups of women including SCD. High BMI also predisposes to early menarche for unclear reasons, but probably due to increased oestrogen levels.[3233] All our patients had a phenotype of HbSS and some reports have suggested that a delay in menarche is more common with the HbSS phenotype than the HbSC phenotype.[34]

Most respondents in our study (82%) had regular menstrual cycles. Samuels-Reid and Scott (1985)[22] however found that menstrual cycles were not as regular in SCD (77%) as compared to controls (95%). However Yoong and Tuck (2002)[35] found much lower figures (30%) of regular cycles among SCD than in our study.

Most respondents (81%) also had normal menstrual cycle lengths of 21–35 days. Samuels-Reid and Scott (1985),[22] documented normal cycle lengths in SCD patients but these were significantly shorter in SCD than controls (mean of 25.6 days in SCD vs. 27.9 days for controls). Yoong and Tuck (2002)[35] found a mean cycle length of 28.9 (±5.95) days, and similar to our study, cycle length was significantly affected by age. Menstrual cycle length is naturally variable towards menarche and menopause which may be influenced by age, because not all cycles are ovulatory.

Most our respondents (97%) had normal menstrual flow duration of 2–8 days. Jimenez et al.[25] reported a mean duration of 3.5 days in SCD versus 4.5 days in controls. Yoong and Tuck (2002)[35] also found a normal duration of flow among SCD patients with a mean duration of menstrual loss was 4.69 (±4.65) days. Still within the normal range, Samuels-Reid and Scott (1985)[22] however found the number of days of flow was significantly longer for the SCD as compared to controls (4.9 vs. 4.3).

Most of our respondents (71%) had light menstrual flow using <3 tampons/napkin per day, and there were no cases of menorrhagia. Yoong and Tuck (2002)[35] also found that 87% of the women with SCD reported either light or moderate menstrual loss, and only 13% had heavy menses or flooding. In a state of chronic anemia, the light menstrual flow may be a protective compensatory mechanism to reduce further red cell loss. On the contrary, however, Samuels-Reid and Scott (1985)[22] found that SCD patients had significantly heavier menses than in the control group (mean number of napkins/tampons used each day of 4.1 in SCD, and 3. 7 in controls). This should be interpreted with care as methods used to assess the quantity of flow was subjective, and available marketed tampons/pads are quite variable in size and absorbent capacity.

In this study, just a little over half of the respondents had menstrual pain, and this was severe in only 35% of cases. Probably because most of our patients had light bleeds and heavier menses are said to be associated with more pain,[2235] as more uterine contractions are required to expel the blood which may be in clots. Djamila et al. (2017)[24] also reported that adolescents with HbSS (SCD) had significantly higher rates of dysmenorrhea than with normal hemoglobin or the SCD trait (63.3% for normal haemoglobin HbAA, 53.6% for the SCD trait HbAS, and 100% for SCD HbSS).

The presence of menstrual pain did not significantly affect the annual frequency of crisis (Chi-square 0.099, df-1, P = 0.753) in our study. Yoong and Tuck[35] found that in 58% of women, crises were associated with menstrual cycles, but this was not very severe. Samuels-Reid and Scott[36] only reported 34% of women having crisis during their menses. Similarly, Sharma et al.[18] demonstrated that acute vaso-occlusive pain was temporally associated with the onset of menstruation in 28% (29 of 103) of women with SCD, and interestingly, that this could be distinguished from dysmenorrhea. In addition to uterine contractions, they postulated several other mechanisms that may explain the association of menstruation to vaso-occlusive crisis; hormonal influences result in a reduction in nitric oxide resulting in endothelial-dependent vasoconstriction and exaggerated pain perception in women with dysmenorrhea.[1837383940]

Few studies have addressed menopause in women with SCD, and our study did not specifically look at postmenopausal SCD patients. It has however been suggested that intravascular sickling, vessel occlusion, and infarction can cause ovarian dysgenesis and premature menopause.[2441] Furthermore, transfusion-related hemochromatosis may be a cause of primary ovarian insufficiency.[24] Five respondents in our study had premature menopause and required further evaluation. A larger study more focused on menopause would be required to make any deductions on this finding.

Study limitations

This was a hospital-based study, so it may introduce some selection bias and not fully represent what may be obtained in the community. Causal relationships are also difficult to determine in cross-sectional studies. Irregular menstrual characteristics may be associated with variable confounding conditions which were not fully accounted for in our study. Menstrual characteristics were self-reported, and this can introduce recall bias.

CONCLUSION

For the study population, the mean age at menarche was 15.6 years, which is slightly high. Most respondents had regular menstrual cycles, normal cycle lengths, normal duration of flow, and light menstrual flow. Overall, 78% of SCD had one or more abnormal menstrual parameters. More than half of the respondents had menstrual pain, and this was severe in 35% of cases, even though 71% of them had light menstrual flow. The presence of menstrual pain did not significantly affect the annual frequency of crisis.

This can serve as a baseline for further larger and follow-up studies, which will be required to validate findings. It will also aid counseling of SCD patients who should be encouraged to keep menstrual diaries.

Financial support and sponsorship

Self-funded.

Conflicts of interest

There are no conflicts of interest.