Plasma Fibrinogen: An Independent Predictor of Pediatric Appendicitis.

AIMS: Appendicitis, in spite of all the diagnostic advances, achieving an accurate and timely diagnosis of this common condition in children remains a challenge. Plasma fibrinogen (FB) is an acute inflammatory mediator and has been proposed and evaluated as an adjunct laboratory marker for improving diagnostic accuracy. The study evaluates the plasma values of Se FB along with other serum markers in pediatric appendicitis patients, to determine their diagnostic accuracy.
METHODS: Prospective observational study on 120 patients between the age group of 5 and 12 years. All eligible enrolled cases underwent total leukocyte count (TLC), plasma FB, C reactive protein (CRP), neutrophil-lymphocyte ratio (NLR), absolute neutrophil count (ANC), and erythrocyte sedimentation rate on admission along with pediatric appendicitis score. Final confirmation of diagnosis and allotment of cohort was made by intra operative findings and histopathological confirmation. Two groups were defined: (1) Histopathologically confirmed acute appendicitis-Cases (2) Nonspecific abdominal pain-Controls. Laboratory results were statistically analyzed between the case and the control groups for diagnostic accuracy.
RESULTS: Study reflected strong statistical significance in terms of leukocyte count, ANC, NLR, CRP, and FB levels. However, plasma FB (value above 4.02 g/L) had the highest diagnostic accuracy rate of 82.50% compared to other laboratory values (TLC-70.83%, CRP-70.00%).
CONCLUSION: Plasma FB has emerged as an accurate diagnostic tool and its diagnostic accuracy is superior to all other laboratory parameter studied (TLC, CRP, NLR, and ANC). Plasma FB values above 4.02 g/L is an independent predictor of appendicitis and can help in reducing negative laparotomy in pediatric age group. Copyright:

INTRODUCTION

Appendicitis remains one of the most common indication for emergent abdominal surgery in pediatric surgery practice and is diagnosed in up to 10% of children evaluated for acute abdominal pain.[1] It presents most frequently in the second decade of life. Complications of the disease is also more common in children younger than 6 years of age, occurring in up to 57% of cases.[2] In spite of all the diagnostic advances; getting a timely and accurate diagnosis for this prevalent condition remains a challenge and many times clinicians depend solely on clinical evaluation. Thus, the decision to operate is about balancing between chances of perforation and negative appendectomy.

There is mounting evidence to suggest that immediate appendicectomy has no special advantage in reducing complications.[3] Deferring surgery in those group of patients with noncontributory lab findings would probably reduce the rate of unnecessary appendicectomies.[3] The present evidence-based recommendation allows for observation and further workup before embarking on operative management, it is at this stage that a more reliable laboratory investigation is required for the diagnosis. Diagnostic accuracy of inflammatory markers to discriminate appendicitis is limited because their levels get elevated in many abdominal pain disorders. Total leukocyte count (TLC), absolute neutrophil count (ANC), neutrophil lymphocyte ratio (NLR), and C - reactive protein (CRP) are the most widely studied and used in clinical practice. However none of them alone could be relied on in isolation, this has led to the formulation of clinical scoring systems like pediatric appendicitis score.[4]

Nonspecific abdominal pain (NSAP) is the most dependable differential diagnosis to be considered in pediatric patients suspected with acute appendicitis (AA).[5] Although a diagnosis of exclusion, it is considered to be a safe diagnosis and so being encountered more frequently in pediatric surgical practice.[6]

The interest in plasma fibrinogen (FB) was started with the study of Menteş et al. in 2012.[7] Later on, seven more studies were reported in the literature; among which only two studies were done exclusively in pediatric population. The aim of our study was to assess the diagnostic accuracy of the biomarkers - plasma FB, along with more traditional markers such as TLC, ANC, CRP, NLR and erythrocyte sedimentation rate (ESR) to diagnose appendicitis in children. Furthermore, to interpret how these investigations can be used effectively to improve the diagnosis and decision making, thus reducing negative appendectomies.

METHODS

This was a prospective observational study on 120 patients between the age group of 5–12 years during the time period from February 2018 to July 2019. They presented with primary complaint of acute right lower abdomen pain to this tertiary pediatric emergency center. The patients on admission were assessed by a single surgeon and were categorized as either clinical AA or NSAP in whom AA was ruled out based on clinical evaluation.

Exclusion criteria included (1) age <5 years, age >12 years, (2) duration of symptoms <6 h, >72 h, (3) concomitant conditions where markers can be elevated (a) hematological, oncological, hepatic dysfunction, infectious, or inflammatory conditions and (b) patients on anticoagulants, anti-inflammatory or antibiotics in the previous month. Ethical clearance of institutional ethical committee was taken and informed consent from the guardian was obtained.

After obtaining history, patients underwent general physical examination and systemic examination. Pediatric appendicitis Score (PAS) as described by Samuel[4] was applied. Hematological examinations (inflammatory markers under study including Plasma FB), biochemistry, urine analysis and ultrasonography (USG) abdomen was performed on admission. The treating surgeon was blinded to all special investigations (FB, CRP, NLR, ANC, ESR) to avoid bias. Clinical appendicitis patients underwent Laparoscopic/open appendicectomy on an emergency basis. Final confirmation of diagnosis was made by intra operative findings and histopathological examination of the postoperative specimen. Only those patients with histological confirmation of appendicular inflammation were included in the appendicitis group (Case group). Negative appendicectomies were considered as errors in diagnosis and included with controls. The NSAP patients were followed up for a period of 2 weeks on outpatient department basis to detect any missed diagnosis of AA before being included in the control group. Laboratory results were later correlated between the case and the control groups for statistical significance.

Statistical analysis

Statistical analysis was done by using data based program software SPSS Version-20, IBM™ (IBM Corp., Armonk, NY, USA). Descriptive statistics such as mean, standard deviation, and percentage were used. Categorical variables are expressed as number of patients and percentage of patients and compared across the groups using Pearson's Chi-square test for Independence of Attributes/Fisher's Exact Test as appropriate. Continuous variables were compared across the groups using unpaired t-test. Receiver operating characteristic (ROC) curve was plotted for each diagnostic variable and sensitivity, specificity, PPV, NPV, and diagnostic accuracy were calculated to understand associations. An alpha level of 5% was taken, i.e., if any P <0.05 it was considered statistically significant. We assessed diagnostic accuracy of inflammatory variables by the area under the curve (AUC) of the ROC curve using the optimal ratio of sensitivity and specificity as the criterion for selecting the best cutoff value corresponding to the variable value with greater discrimination power between different groups.

RESULTS

A total of 120 patients were enrolled in the study with initially 60 patients in each group. Out of the 60 patients operated with clinical diagnosis of appendicitis, 55 patients were finally included in the case group taking into account the peroperative finding and histopathological confirmation. The rest 5 patients were found to have normal appendix histopathologically, so were included in the control group for statistical analysis. None in the clinical NSAP group during follow-up period exhibited signs of a missed AA.

The two groups were comparable statistically in terms of age and demographic variables. Descriptive and univariate analysis of all variables is shown in Table 1. The Study reflected strong statistical significance in terms of leukocyte count, ANC, NLR, CRP, and FB levels. ESR was the only studied variable which lacked statistical significance. The diagnostic accuracy of each of the statistically significant variable was further evaluated for positive predictive value, negative predictive value, sensitivity, specificity and diagnostic accuracy [Table 2]. TLC which is the most commonly used diagnostic value had a moderate accuracy rate of 70.83 (value above 14,000/mm3), whereas plasma FB presented a higher diagnostic accuracy rate of 82.50 with an especially high sensitivity rate of 81.81% (value above 4.02 g/L) [Table 4].

Table 1

Descriptive and univariate analysis of inflammatory markers

	True appendicitis group(55)			Control group (65)			P	Significance
	Mean	Median	SD	Mean	Median	SD
Age	8.52	8.00	2.40	8.72	9.00	2.07	0.611	Not significant
TLC/mm3	14,770.52	14,050.00	5177.95	9987.50	10,250.00	3409.63	<0.001	Significant
ANC (cell/µl)	10,899.79	10,375.50	8645.01	6272.77	5349.00	3033.25	<0.001	Significant
NLR	4.78	3.27	3.70	2.71	2.00	2.10	0.001	Significant
ESR (mm/1st h)	25.93	24.00	12.54	24.65	25.00	11.39	0.661	Not significant
CRP (mg/L)	39.81	32.03	28.22	24.65	10.00	23.42	<0.001	Significant
FB (mg/dl)	482.41	501.30	146.76	308.59	293.50	122.72	<0.001	Significant

SD: Standard deviation, TLC: Total leukocyte count, ANC: Absolute neutrophil count, NLR: Neutrophil-lymphocyte ratio, ESR: Erythrocyte sedimentation rate, CRP: C reactive protein, FB: Fibrinogen

Table 2

Diagnostic accuracy of inflammatory markers

Parameters	TP	TN	FP	FN	Sensitivity	Specificity	PPV	NPV	Diagnostic accuracy
TLC (/mm3)	25	60	5	30	45.45	92.30	83.33	66.66	70.83
ANC (cell/µl)	28	57	8	27	50.09	87.69	77.78	67.85	70.83
NLR	33	47	18	22	60.0	72.30	64.70	68.11	66.67
CRP (mg/L)	49	35	30	6	89.90	53.84	62.02	85.36	70.00
FB (mg/dl)	45	54	11	10	81.81	83.07	80.35	84.34	82.50

TLC: Total leukocyte count, ANC: Absolute neutrophil count, NLR: Neutrophil-lymphocyte ratio, CRP: C reactive protein, FB: Fibrinogen, PPV: Positive predictive value, NPV: Negative predictive value, TP: True positive, TN: True negative, FP: False positive, FN: False negative

Table 4

Comparison of plasma fibrinogen in appendicitis (all studies)

	Sensitivity (%)	Specificity (%)	Predictive value of positive test (%)	Cutoff (mg/dl)
Present study [Reference]	81.81	83.07	82.5	402.5
Menteş et al.[7]	70.39	50	91.97	>245.5
Zhao et al.[23]	83	70	73	>400
Alvarez-Alvarez et al.[24]	88.24	89.36	92.31	896
Ulukent et al.[30]	57	59	60.4	300
Nyuwi et al.[31]	88	55	-	375
Makadia and Jain[32]	81	60	96	>300
Prada-Arias et al.[33]	43.9	79.7	-	510

ROC curve analysis as summarized in Figure 1 and Table 3, illustrates that FB value had the highest AUC giving the highest specificity and sensitivity among all the individual variables and closest to the PAS score (score above 4.5).

Figure 1

Receiver operating characteristic curve of appendicitis group and control group

Table 3

Area under the curve of inflammatory markers in study group

Test result variable (s)	AUC	Cut off value	P
TLC/mm3	0.777	14,930.00	<0.001
ANC (cell/µl)	0.749	10,485.50	<0.001
NLR	0.677	2.9	0.001
ESR (mm/1st h)	0.529	-	0.588
CRP (mg/L)	0.681	10.50	0.011
FB (g/l)	0.824	0.4025	<0.001
PAS	0.941	4.5	<0.001

TLC: Total leukocyte count, ANC: Absolute neutrophil count, NLR: Neutophil-lymphocyte ratio, ESR: Erythrocyte sedimentation rate, CRP: C reactive protein, FB: Fibrinogen, PAS: Paediatric appendicitis score, AUC: Area under the curve

DISCUSSION

The goal of modern surgical sciences is to attain an acceptable balance between complications of the disease versus complications of the treatment. In the past, appendicitis has been considered a surgical emergency that requires prompt appendectomy to avoid perforation and other complications. Perforation occurs rarely in the first 12 h of symptoms but is more likely with passage of time thereafter, becoming common after 72 h.[3] The argument that there is only negligible operative morbidity associated with negative laparotomy, to allow the risk of perforation is also being proved wrong.[8] Thus the practice of rushing into surgery is no longer advocated, observation and investigation is becoming more and more acceptable to allow for more definite diagnosis before surgical exploration.

A number of serum markers, imaging modalities and clinical scoring system were introduced to reduce negative appendicectomy rates; however, their usefulness in clinical practice has not yet been established. White blood cells (WBC) count is probably the only routine test performed in clinical practice, but its elevation can be highly nonspecific affected by wide spectrum of conditions. Various studies[910] in large populations found unacceptably low specificity and sensitivity to depend on this parameter in isolation. The popular scoring system ALVARADO score had shown better promise but the accuracy has been reported to range from 50% to 95%.[911]

Plasma FB, is an acute inflammatory mediator, and its plasma level is usually raised in any acute inflammatory condition and so is expected to rise in AA also. FB or clotting factor I is a plasma glycoprotein, apart from being a key factor in hemostasis it has important pro-inflammatory functions associated with its ability to bind to receptors and activate several types of immune cells involved in the inflammatory response.[12] This fibrinopeptide can operate as a chemoattractant and has the potent activity to recruit both polymorphonuclear cells (PMN) and fibroblast. They are involved in the signaling and activation of mononuclear phagocytes from altered coagulation or fibrinolysis in an inflamed tissue or organ.[13] The effects on leukocyte activation is mediated through specific leukocyte receptors, the intern receptor αMβ2.[1415] The normal plasma values of FB range between 2 and 4 g/L with a plasma half-life of 4 days, however in conditions associated with infection, injury and inflammation marked with vascular disruption the values rises many times.[1617]

FB signaling through CD11b/CD18 during acute Inflammation lead to local production of inflammatory cytokines, such as tumor necrosis factor-a and interleukin-1b.[181920] These mediators were significantly elevated with diagnostic utility in AA as indicated by Sack et al. in 2006.[21] Further Li et al. in 2011 studied endotoxin-induced activation of the extrinsic coagulation pathway in patients with AA and thus FB was expected to be utilized as a potential marker for predicting appendiceal perforation.[22]

The first study to evaluate diagnostic relevance of se FB in AA was done by Menteş et al. in 2012.[7] The study involved 201 patients and found a positive correlation, the best cut off value of FB was above 2.455 g/L, with a sensitivity of 70.39%, specificity of 50% and positive predictive value of 91.97%. Zhao et al. who studied WBC, D-dimer and FB level in 455 confirmed adult cases of appendicitis found that specificity of FB for the preoperative diagnosis of appendiceal perforation was higher (0.70) compared with WBC (0.35) and D-dimer (0.68). The optimal ratio of sensitivity (0.87) and specificity (0.71) was calculated when the FB level was >4.0 g/L for perforated appendicitis.[23] Averez studied the variation in FB levels among uncomplicated and complicated appendicitis and inferred that higher FB cutoffs at 885 mg% would be highly accurate to detect complicated appendicitis (sensitivity - 85.9% and specificity-91.49%) it is noteworthy that in the same study, even among uncomplicated appendicitis the mean values were significantly high at 6.754 ± 2.134 g/L indicating its utility in diagnostic role of appendicitis in general.[24]

Most of these studies available were in adults except for two. The diagnostic cut off are likely to be different in the pediatric population and require separate study which was the principle aim for the present study. Pediatric AA remains more elusive in diagnosis as well as more prone to complications. Negative appendicectomy rate in children is around 30%–46% as compared to <20% in adult patients.[92526]

The retrospective study by Feng et al. involving 466 pediatric patients with histologically confirmed appendicitis concluded that children with hyperfibrinogenemia and clinical symptoms of appendicitis may be regarded as at higher risk of appendiceal perforation than whose FB level is normal.[27] The specificity and positive predictive value of FB were considerably higher in comparison with WBC or CRP in predicting perforation. The study also tried to define a cut off value for optimum specificity and sensitivity through an Relative Operating Levels (ROL) curve plotting. At a value of 5 G/DL the specificity and sensitivity were 0.74 and 0.82, respectively. Comparatively, the specificities of leukocytosis (0.25) and CRP (0.34) were much lower in predicting perforation.

There is only one study in patients below 5 years, the age group were misdiagnosis rates can go as high as 70%.[228] This study by Prada-Arias et al. studied [Table 4] 82 patients below 5 years with clinical diagnosis of AA out of which there were 55 had confirmed AA (17 uncomplicated and 38 complicated) and 27 NSAP. Plasma FB not only had good accuracy in differentiating between NSAP and uncomplicated AA (Acc- 0.77) but also in differentiating complicated versus uncomplicated AA (Acc- 0.73).[29]

There is one study which has failed to demonstrate any positive result [Table 4], this study by Ulukent et al.[30] compared various factors including FB in 197 adult patients who underwent surgical exploration for clinical diagnosis of AA, out of the 197 cases operated 28 laparotomies were negative (14.2%). Results of this study were contrary to other studies, The mean FB levels were higher (3.49 ± 0.18 g/L) in negative appendicectomy group than in histologically positive group (3.05 ± 0.07 g/L).[30] A similar study design by Nyumi [Table 4] comparing FB levels between histologically proven AA and negative appendicectomy patients, FB level was significantly higher in positive histology group (4.36 ± 0.40 g/L vs. 3.91 ± 0.66 g/L) and suggested that a cutoff value of 4.00 g/L will be ideal for diagnosis.[31]

Combination of more than one diagnostic criteria can be an alternative to achieve better positive diagnostic rates; for example a combination of PAS score and USG abdomen along with plasma FB was shown to have far superior accuracy in a study done by Makadia and Jain on 213 patients [Table 4] operated for AA.[32]

Most studies confirm positive correlation of plasma FB level with severity of the disease, so can be especially useful in differentiating complicated from uncomplicated AA. With growing evidence suggesting that select uncomplicated AA can be managed nonoperatively, these laboratory test can be utilised to predict onset of complications requiring intervention.

Our study has used prothrombin time derived estimation by photo-optical Coagulometer for FB analysis, the other method available is the Clauss method. Variability in the two methods prevents establishing a common reference range between the two, so it becomes pertinent that the method used is specified for analyzing the values in research and clinical practice. Among the two methods prothrombin time derived estimation is more useful in hyperfibrinogenic states requiring quantitative more than qualitative assessment. The test is relatively cheap (costing approximately Rs. 300) and ideal for its routine use in clinical practice.

CONCLUSION

Appendicitis is associated with a systemic inflammatory response. The present research investigated elevated plasma FB as a predictive marker for appendicitis. It has been found that plasma FB is more accurate in diagnosing appendicitis and impending perforation, compared to other inflammatory markers. The univariate analysis showed TLC, ANC, NLR, and CRP were significantly raised in appendicitis. However, they have lower potential to aid the preoperative diagnosis of appendicitis compared to plasma FB. This study showed that plasma FB is an independent predictor of appendicitis and can be helpful in reducing negative appendectomy rates in children.

Financial support and sponsorship

Viz., grants, equipment, drugs or all of these: Government Institution.

Conflicts of interest

There are no conflicts of interest.