The diagnostic and discriminatory value of neutrophil lymphocyte ratio (NLR) in acute appendicitis

ABSTRACT

Introduction: White cell count (WCC) and C-reactive protein (CRP) are of limited value in the early detection of acute appendicitis. The neutrophil lymphocyte ratio (NLR) has been proposed as a novel and cost-effective marker that could play this role.

Aim: To evaluate the usefulness of neutrophil lymphocyte ratio in the diagnostic work-up of acute appendicitis in a South African setting.

Methods: Retrospective review of patient records who were treated for acute appendicitis over 13.5 years was conducted. Data retrieved included patients’ demography, pre-operative WCC with differential count, CRP and histology result. Sensitivity, specificity, positive predictive value and negative predictive value of WCC, CRP and NLR were calculated. Receiver operating characteristic (ROC) curves were generated by mapping true positives (sensitivity) against false positives (1-specificity) to establish cut-off values.

Results: A total of 1098 records were found of which 590 were excluded due to incomplete data. There were 56.7% males with an average age of 28 years. Sensitivity, specificity, positive and negative predictive values of the NLR for diagnosis of acute appendicitis at a cut-off value of 2.6 were 81.1%, 47.5%, 94.4% and 18.8% respectively. For WCC at cut-off value of 9.2 the corresponding findings were 67.9%, 55.0%, 94.4% and 13.7%; respectively. Corresponding results for CRP of 13 were 89.8%, 42.5%, 94.4% and 27.9%, respectively. An NLR of 4.9 differentiated complicated from uncomplicated acute appendicitis. The AUC in the ROC curves for NLR, WCC and CRP were 0.662, 0.622 and 0.687; respectively.

Conclusion: The neutrophil to lymphocyte ratio was not useful in the diagnosis of uncomplicated acute appendicitis in our setting but could differentiate complicated from uncomplicated acute appendicitis.

Main article text

Introduction

Acute appendicitis is the commonest cause of acute abdomen requiring emergency surgery. Although the diagnosis of acute appendicitis is mainly clinical, diagnosis is not always straight forward as its presentation is classical in less than half of the patients (1). Consequently, the diagnosis of acute appendicitis is sometimes made late which increases the risk of complications such as rupture with generalized peritonitis (1–3).

Frequently performed laboratory tests such as the white cell count (WCC), pro-calcitonin and C-reactive protein (CRP) levels are not useful for early diagnosis of acute appendicitis i.e. before it has complicated (4–8). Reliability of available laboratory tests for acute appendicitis is influenced by the initiating process i.e., whether it was obstructive or inflammatory and the time which would have elapsed since the onset of the pathological process (4).

In patients who have an acute inflammatory process, a combination of de-margination, delayed apoptosis and increased proliferative activity of stem cells in the bone marrow leads to neutrophilia. Concurrently, lymphocytopaenia develops due to enhanced margination and accelerated apoptosis of lymphocytes. The resulting neutrophilia and lymphocytopaenia leads to an increase in neutrophil to lymphocyte ratio (NLR) (9).

The NLR has been proposed as a novel, cost effective and easily available inflammatory marker for supporting diagnosis of acute appendicitis (10–14). It is reported to be highly sensitive and specific for the diagnosis of acute appendicitis as compared to WCC and CRP (14–16).

Normal values of haematological parameters are influenced by race, gender and geography (16–18). The value of WCC, CRP and NLR has thus far not been studied in South Africa and thus the aim of this study was to evaluate the reliability of NLR as compared to WCC and CRP for diagnosis of acute appendicitis in a South African setting. Furthermore, the ability of NLR to distinguish simple from complicated acute appendicitis was also studied.

Method

A retrospective review of records of patients 10 years and older who had appendicectomy from January 2003 to June 2015 was conducted at Chris Hani Baragwanath Academic Hospital, Charlotte Maxeke Johannesburg Academic Hospital, Helen Joseph Hospital and Leratong Hospital. Ethical permission to conduct the study was received from the relevant institutional authorities. Data retrieved included patients’ demography, pre-operative full blood count (FBC), CRP and histology results. Only patients, for whom results for WCC with differential count and CRP could be traced, were included in the study. Reference ranges for normal WCC and CRP used in this study were 4-11 × 10⁹/l and below 10 mg/l, respectively. For each patient NLR was calculated manually.

Retrieved records were divided into four groups based on histology results. The groups were: normal appendix (Group 1), simple or catarrhal acute appendicitis (Group 2), complicated acute appendicitis i.e. suppurative or gangrenous (Group 3) and serosal appendicitis (Group 4). Serosal appendicitis meant that the outer layer of the appendix had inflammatory changes due to contiguous spread from an adjacent organ or generalized peritonitis from another source.

Collected data was entered into excel spreadsheet and statistical analysis was performed using XLSTAT. The mean and standard deviation (SD) and median with range were calculated. Sensitivity, specificity, positive predictive value (PPV) and negative predictive values (NPV) for WCC, CRP and NLR were calculated for the four groups of patients based on histological findings. Diagnostic value of each test and final histology were compared using a Chi-square test and a p-value less than 0.05 was considered statistically significant.

Receiver operating characteristic (ROC) curve was generated by plotting true positives (sensitivity) along the Y-axis against false negatives (1-specificity) along the X-axis for NLR, WCC and CRP for diagnosis of simple and complicated appendicitis. Area under the curve (AUC) with 95 percent confidence interval was computed for each of the diagnostic tests. The cut-off values for WCC, CRP and NLR were derived from ROC curves i.e. the most prominent point towards top left hand corner on the curve. The ROC curves and corresponding AUC were used to compare the usefulness of WCC, CRP and NLR in the diagnosis of acute appendicitis and for separation of complicated from uncomplicated acute appendicitis (19,20). Based on the AUC for each; WCC, CRP and NLR were considered perfect (AUC = 1.0), excellent (AUC = 0.9-0.99), good (AUC = 0.8-0.89), fair (AUC = 0.7-0.79) and not useful (AUC < 0.7) to support the diagnosis of acute appendicitis (21,22).

Results

A total of 1098 patients had appendectomy during the study period of which 590 were excluded due to incomplete data. Males constituted 56.7% of the study population and the average age of included participants was 28.0 years (range: 18-60). 81.5% (414/508) of the patients were 40 years or younger (Table 1).

Table 1:

Demographic parameters (N= 508)

Parameter	Number	Percentage
Total records	1098	100%
Exclusion	590	53.7%
Inclusion	508	46.3%
Gender
Males	288	56.7%
Females	186	36.6%
Not specified	34	6.7%
Age groups
<20 years	187	36.8%
20-40 years	227	44.7%
41-60 years	75	14.8%
>60 years	19	3.7%

In 74.4% (378/508) specimens, histology showed complicated acute appendicitis and in 7.9% (40/508) of specimens normal appendices were reported (Table 2).

Table 2:

Histopathological findings (N=508)

Histological finding	Number	Percentage
Group 1	40	7.9%
Group 2	55	10.8%
Group 3	378	74.4%
Group 4	35	6.9%
Total	508	100%

Group 1= Normal appendix, Group 2= Simple acute appendicitis, Group 3= Complicated acute appendicitis, Group 4= Serosal appendicitis

The rate of negative appendicectomy (normal appendix on histology) for men was 6.3% (18/288) compared to 10.2% (19/186) for women. 81.3% (234/288) of men had complicated acute appendicitis as opposed to 62.4% (116/186) in women (Table 3).

Table 3:

Histopathological findings acccording to gender

Histological group	Males (n= 288)	Females (n= 186)
Group 1	18 (6.3%)	19 (10.2%)
Group 2	25 (8.7%)	29 (15.6%)
Group 3	234 (81.3%)	116 (62.4%)
Group 4	11 (3.8%)	22 (11.8%)

The mean±SD of NLR for patients who had negative appendicectomy and simple acute appendicitis were 6.8±7.0 and 10.4±10.9, respectively. For complicated acute appendicitis the mean NLR was 12.1± 11 (Table 4).

Table 4:

Mean and standard deviation of WCC, CRP and NLR of pathological groups

Test	Group 1	Group 2	Group 3	Group 4
WCC	12.4 ± 10.4	13.4 ± 7.2	14.4 ± 8.7	14.7 ± 9.8
CRP	84.7 ± 107.0	84.0 ± 88.5	171.3 ± 142.4	186.7 ± 142.7
NLR	6.8 ± 7.0	10.4 ± 10.9	12.1 ± 11	9.9 ± 6.4

The cut-off NLR as per ROC curve was 2.9. Cut-offs values for CRP and WCC were 13 g/dl and 9.2 × 10⁹/l, respectively. The overall sensitivity of NLR to support diagnosis of acute appendicitis for NLR was 81.1%, for CRP 89.8% and 67.9% for WCC at the already specified cut-off points (Table 5).

Table 5:

Sensitivity, specificity, PPV and NPV of cut off values of WCC, CRP and NLR of groups

Test	Diagnosis	Cut off value	Sensitivity	Specificity	PPV	NPV	p-Value
WCC	Group 2 + 3	9.2	67.9%	55.0%	94.4%	13.7%	0.011
	Group 4	9.2	68.6%	55.0%	57.1%	66.7%	0.091
CRP	Group 2 + 3	13	89.8%	42.5%	94.4%	27.9%	<0.0001
	Group 4	90.7	68.6%	70.0%	66.7%	71.8%	<0.0001
NLR	Group 2 + 3	2.6	81.1%	47.5%	94.4%	18.8%	0
	Group 3	4.9	73.5%	68.9%	90.3%	20.0%	0.08
	Group 4	10.1	54.3%	82.5%	73.1%	67.3%	0.004

Group 2 + 3= (simple and complicated acute appendicitis), PPV= positive predictive value, NPV= negative predictive value.

The area under the ROC curve (AUC) for NLR was 0.662 (95% CI 0.578-0.750) with a p-value of 0.000. For WCC the AUC of ROC was 0.622 (95% CI 0.528-0.715; p-value of 0.011). The AUC for CRP was 0.687 (95% CI 0.598-0.777; p<0.0001).

Discussion

Amongst key findings from this study is that acute appendicitis is more common in males and more than 70% of patients present when the disease is already complicated. The above is consistent with findings from previous studies (1–3,23). Negative appendicectomy rate and serosal appendicitis were reported more in females than males, which is similar to what was found by Sammalkorpi et al (24) and Pranesh et al (25). An overall negative appendicectomy rate below 10% in our environment (a medium to low-income setting), is worrying because a complicated acute appendicitis rate of 74.4%, suggests either that the patients presented late (after an average 72 hours since the onset of symptoms) or the diagnosis of acute appendicitis in majority of patients was delayed until the clinical signs became classical.

Conflicting results have been reported regarding the usefulness of laboratory parameters as adjuncts for diagnosis of acute appendicitis. Yu et al (4), Bates et al (8) and Ebied et al (26) found that the WCC was the least helpful in supporting diagnosis of both simple and complicated acute appendicitis. NLR in the current study was more sensitive but less specific compared to WCC for diagnosis of acute appendicitis. On the other hand, when compared to CRP, NLR was more specific but less sensitive for diagnosis of acute appendicitis which is contrary to findings from a study conducted by Kim et al (27). Panagiotopoulou et al (7) and Qi et al (28) also found that CRP was more reliable in supporting the diagnosis of acute appendicitis, especially if the disease is complicated disease. The superiority of CRP over other laboratory tests in the diagnostic work-up of patients suspected to be having acute appendicitis was also confirmed in a meta-analysis conducted by Yu et al (4).

That CRP would be more useful as compared NLR in the diagnostic investigation of patients who are suspected of having acute appendicitis is contrary to reports by Yazici et al (12), Ishizuka et al (29) and Kucuk et al (14). Finding of superior diagnostic value of CRP over NLR in the current study may be due to high prevalence of complicated acute appendicitis in the cohort of patients studied. Unlike the neutrophilic response which is quick and stimulated almost immediately by cytokines, CRP is a protein and therefore has to await synthesis by the liver. It makes sense CRP would be more reliable and is therefore likely to be raised in complicated acute appendicitis, which in most cases is a time dependent phenomenon (4). The mean CRP was also the highest in patients who had serosal appendicitis, which might have been because the underlying conditions were severe and longstanding.

A cut-off value of NLR of 2.6 obtained in the current study was higher than the 1.73 used by Kucuk et al (14). However, despite a lower cut-off value their study found NLR useful in the diagnostic investigation of suspected acute appendicitis. The above is despite them finding the AUC on the ROC curve below 0.8 i.e. 0.6. Actually, an overwhelming majority of the studies which reported NLR as being reliable as an aid for diagnosis of acute appendicitis did so even when the AUCs obtained were far below 0.8 (10–14). If recommendations as advised by Carter et al were followed, the NLR would have been found to be of no added value in the diagnostic work-up of acute appendicitis, as it was in the present study (21). Explanation as to when the AUC should be considered useful has also been well explained by Park et al in 2004 (30). However, as demonstrated in the current study, NLR can discriminate uncomplicated from complicated acute appendicitis.

Various factors may affect and therefore influence the NLR. These factors include the normal baseline haematology values (16–18), co-morbid conditions (31,32) and stage of inflammatory process at presentation of patients who have acute appendicitis (33–35). It is worth noting that communicable diseases are highly prevalent in South Africa, which may affect baseline neutrophil and lymphocyte counts. These conditions include human acquired immunodeficiency virus (HIV) infection which may lead to depletion of CD4⁺ T-lymphocytes (34) and therefore affect the diagnostic value of NLR.

Another plausible explanation of why NLR was not useful in the South African setting may be because of regional variations in baseline value of haematological parameters (16–18). Bain et al (16) found that WCC and absolute neutrophil count to be significantly lower in black patients. The same group however found that the lymphocyte count is not influenced by race. The above would be significant as a low baseline lymphocyte count would result in a lower pre-morbid NLR. Majority of patients in the current study were most probably blacks, which is a plausible explanation for the low NLR.

An additional finding from this study is that NLR was able to distinguish between complicated and uncomplicated acute appendicitis. An NLR cut-off of 4.9, differentiated complicated from uncomplicated acute appendicitis. Other studies have found NLR to have similar capabilities. However CRP was found to be more helpful in supporting both simple and complicated acute appendicitis in the current study unlike what was reported previously. Our findings are not surprising as an overwhelming majority of patients had complicated acute appendicitis. C-reactive protein level is known to be less helpful in the early stage of acute appendicitis i.e. less than 12-24 hours. The NLR was very high for the serosal acute appendicitis group (10.1), pointing to the possibility of a severe inflammatory process in this group.

Limitations

More than 50% of the records were excluded because they did not have a complete set of results. The time at which blood tests were done relative to appendicectomy was unknown. Co-morbid conditions such as HIV and TB which are prevalent in our setting and could have affected the NLR were not recorded. The sample size for serosal appendicitis was very small to allow for a valid comparison. As the yardstick for the study was the final histology result and therefore dependent on the expertise of pathologists, it is probably that the analysis could have been affected by verification bias. Historically, different cut-off values of NLR are used.

Conclusions

Neutrophil to lymphocyte ratio was higher in complicated compared to uncomplicated acute appendicitis. The purported usefulness of NLR reported in some previous studies may have been due to misunderstanding of ROC curves which made values below 0.8 significant. Diagnosis of acute appendicitis should remain clinical, especially in low income settings where majority of patients present late when the disease has already complicated and the usefulness of NLR is surpassed by both clinical findings and level of CRP. Further studies are suggested to verify the purported usefulness of NLR in the diagnostic work-up of acute appendicitis and other acute medical emergencies.

Acknowledgements

We would like to thank Mr. Tinyiko Ngobeni and Mrs. Sue Candy from the National Health Laboratory Services.

References

1. NshutiR, KrugerD, LuvhengoTE. Clinical presentation of acute appendicitis in adults at the Chris Hani Baragwanath academic hospital. International Journal of Emergency Medicine 2014; 7:12

2. EdinoST, MohammedAZ, OchichaO, AnumahM. Appendicitis in Kano, Nigeria: A 5-year review of pattern, morbidity and mortality. Ann Afr Med 2004; 3(1):38 – 41.

3. KongVY, BulajicB, AllortoNL, HandleyJ, ClarkeDL. Acute appendicitis in a developing country. World J Surg 2012; 36(9):2068–2073

4. YuCW, JuanLI, WuMH, ShenCJ, WuJY, LeeCC. Systemic review and meta-analysis of the diagnostic accuracy of procalcitonin, C-reactive protein and white blood cell count for suspected acute appendicitis. BJS 2013; 100: 322–329

5. XharraS, Gashi-LuciL, XharraK, VeselajF, BicajB, SadaF, et al. Correlation of serum C-reactive protein, white blood count and neutrophil percentage with histopathology findings in acute appendicitis. World J Emerg Surg 2012; 7:27

6. JamaluddinM, HussainSMA, AhmadH. Acute appendicitis with normal total leukocyte count. Journal of Surgical Academia 2013; 3(1): 2–6.

7. PanagiotopoulouIG, ParasharD, LinR, AntonowiczS, WellsAD, BajwaFM, KrijgsmanB. The diagnostic value of white cell count, C-reactive protein and bilirubin in acute appendicitis and its complications. Ann R Coll Surg Engl 2013; 95: 215–221

8. BatesMF, KhanderA, SteigmanSA, TracyTF, LuksFI. Use of White Blood Cell Count and Negative Appendectomy Rate. PEDIATRICS 2014; 133 (1): e39–44.

9. ZahorecR. Ratio of neutrophil to lymphocyte counts - rapid and simple parameter of systemic inflammation and stress in critically ill. BMJ 2001; 102(1):5–14.

10. PrasetyaD and GunadiR. Accuracy of neutrophil lymphocyte ratio for diagnosis of acute appendicitis children; A diagnostic study. Annals of Medicine and Surgery 2019; 48: 35–38

11. MarkarSR, KarthikesalingamA, FalzonA, KanY. The diagnostic value of neutrophil:lymphocyte ratio in adults with suspected acute appendicitis. Acta Chir Belg 2010; 110(5): 543–547

12. YazıcıM, ÖzkısacıkS, ÖztanMO, GürsoyH. Neutrophil/lymphocyte ratio in the diagnosis of childhood appendicitis. The Turk J of Pediatr 2010; 52(4): 400–403

13. KahramancaS, ÖzgehanG, ŞekerD et al. Neutrophil-to-lymphocyte ratio as a predictor of acute appendicitis. Ulus Travma Acil Cerr Derg 2014; 20(1):19–22.

14. KucukE. The change of neutrophil-lymphocyte ratio in acute appendicitis. Med-Science 2015, 4(3):2379–2387

15. An appraisal of inflammatory markers in distinguishing acute uncomplicated and complicated appendicitis. Dig Dis 2016; 33: 177–181

16. BainB, SeedM,GodslandI. Normal values for peripheral blood white cell counts in women of four different ethnic origins. J. Clin. Pathol 1984; 37: 188–193

17. MalikMO. Normal reference value of blood cell count, red, white and platelet of Khartoum North Area. Al Neelain Medical Journal 2013; 3(8): 100–109

18. TembeN, JoaquimO, AlfaiE, et al. Reference values for clinical laboratory parameters in young adults in Maputo, Mozambique. PLoS ONE 2014; 9(5): e97391

19. SwetsJA. Measuring the accuracy of a diagnostic system. Science 1988; 240(4857): 1285–1293

20. ZweigMH, CampbellG. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993; 39(4); 561–577

21. CarterJV, PanJ, RaiSN, GalandiukS. ROC-ing along: evaluation and interpretation of receiver operating characteristic curves. Surgery 2016; 159: 1638–1645

22. PerkinsNJ & SchistermanEF. The Inconsistency of “Optimal” Cut-points Using Two ROC Based Criteria. Am J Epidemiol. 2006; 163(7): 670–675

23. TuckerA, SloanK, GartsinI, VerghisR. White cell counts, CRP and appendicitis – is there a role for pre-operative blood tests? A cohort study. Journal of Health and Medical Informatics 2015; 6: 2 [[Cross Ref]].

24. SammalkorpiHE, MentulaP, LeppäniemiA. A new adult appendicitis score improves diagnostic accuracy of acute appendicitis - a prospective study. BMC Gastroenterol 2014; 14: 114. [[Cross Ref]].

25. PraneshN, SathyaV, MainprizeKS, Serosal appendicitis: incidence, causes and clinical significance. Postgrad Med J 2006; 82: 830–832

26. EbiedEF, EbiedH. The diagnostic value of C-reactive protein and white blood cell count in diagnosis of acute appendicitis. Egypt J Surg 2016; 35(1): 1–4.

27. KimE, SubhasG, MittalVK, GolladayES. C-reactive protein estimation does not improve accuracy in the diagnosis of acute appendicitis in pediatric patients. Int. J Surg 2009; 7:74–77.

28. QiFQ, ZhangB. Clinical significance of C-reactive protein levels in the determination of pathological type of acute appendicitis. Int J Clin Exp Med 2015; 8(8):13887–13890.

29. ParkSH, GooJM, JoCH. Receiver Operating Characteristic (ROC) Curve: Practical Review for Radiologists. Korean J Radiol 2004; 5(1): 11–18.

30. IshizukaM, ShimizuT, KubotaK. Neutrophil-to-Lymphocyte ratio has a close association with gangrenous appendicitis in patients undergoing appendectomy. Int Surg 2012; 97(4):299–304.

31. Vaughan-ShawPG, ReesJRE, KingAT. Neutrophil lymphocyte ratio in outcome prediction after emergency abdominal surgery in the elderly. Int. J Surg 2012; 10: 157–162.

32. SalciccioliJD, MarshallDC, PimentelMAF et al. The association between the neutrophil to lymphocyte ratio and mortality in critical illness: an observational cohort study. Crit. Care 2015; 19:13

33. SoldoI, BiljakVR, BakulaB, BakulaM, SimundicAM. The diagnostic accuracy of clinical and laboratory parameters in the diagnosis of acute appendicitis in the adult emergency department population – a case control pilot study. Biochem Med (Zagreb) 2018; 28(3): 030712

34. DjordjevicD, RondovicG, SurbatovicM, StanojevicI, UdovicicI, et al. Neurtophil-to-Lymphocyte Ratio, and mean Platelet Volume-to-Platelet Count Ratio as Biomarkers in Critically Ill and Injured Patients: Which Ratio to Choose to Predict Outcome and Nature of Bacteremia. Mediators Inflamm. 2018;3758068 [Cross Ref].

35. BrenchleyJM, SchackerTW, RuffLE, PriceDA, TaylorJH, BeilmanGJ, et al. CD4 ⁺T cell depletion during all stages of HIV disease occurs predominately in the gastrointestinal tract. J. Exp. Med. 2004; 200(6): 749–759

Section