Table 2.

Prognostic estimates effect of NLR and conclusion of each study

Author	CURB‐65 (2–5)	End point	Cutoff of NLR	Sensitivity	Specificity	Prognostic effect estimates (HR, OR)	AUC	Variables	Conclusion
de Jager et al. 13	123	Adverse event (mortality/adverse events)	10	74 (95% CI: 59.66–85.37)	53.33 (95% CI: 47.91–58.69)	N/A	N/A	Age, gender, comorbidities, medications, pathogens, CRP level	NLR better‐predicted mortality compared to CRP levels, WBC count, neutrophil count, and lymphocyte count
		In‐hospital mortality	10	78.26 (95% CI: 56.3–92.54)	51.61 (95% CI: 46.4–56.8)	N/A	0.701
Avci and Perincek 14	N/A	30‐day mortality	N/A	N/A	N/A	N/A	0.577 (95%CI 0.501‐0.650)	Age, smoking, comorbidities, complication, clinical parameters	NLR showed low 30‐day mortality estimation accuracy than PSI class, PSI scores and procalcitonin
Cataudella et al. 15	175	30‐day mortality	11.2	100 (95% CI: 92.45–100)	77.7 (95% CI: 70.14–84.13)	N/A	0.94	Age, sex, CURB‐65, PSI, comorbidities	NLR predicted mortality better than PSI, CURB‐65, CRP, and WBC count
			13.4	91.49 (95% CI: 76.84–89.33)	83.78 (95% CI: 76.84–89.33)	N/A
Curbelo et al. 16	128	30‐day mortality	10	63.6 (95% CI: 35.4–84.3)	65 (95% CI: 56.8–72.4)	OR = 1.04 (1.0–1.1)	0.88 (95% CI: 0.79–0.98)	Age, sex, vaccination, comorbidities, PSI, CURB‐65, CRP, procalcitonin, proadrenomedullin	NLR was not inferior to proadrenomedullin and significantly better than other biomarkers
Masbang et al. 17	N/A	Predicting HR from LR and MR	10.24	56.4	66.8	N/A	0.726	Age, gender, comorbidity, smoking, symptoms, vital signs, radiologic finding	NLR predicts CAP severity more than WBC count. NLR can better predict HR from LR and MR
Ozmen et al. 18	N/A	30‐day mortality	N/A	N/A	N/A	HR = 1.04 (0.99–1.01)	0.64 (0.49–0.79)	Age, gender, comorbidity, smoking, biochemistry, ventilation, ABG finding	Higher NT‐pro BNP values (above 2000 pg/ml) and NLR can be used to predict pneumonia severity and higher NLR on admission to ICU has a higher risk of 180‐day mortality
		Mortality 180 days after ICU admission	N/A	N/A	N/A	HR = 1.04 (1.01–1.07)	0.63 (0.52–0.74)
		ICU mortality	N/A	N/A	N/A	N/A	0.60(0.46–0.75)
Postma et al. 12	N/A	30‐day mortality	Median = 10.4	N/A	N/A	OR = 1.19 (95% CI: 1.02–1.38)	N/A	Age, sex, comorbidities, clinical parameters, mortality scores	NLR had a moderate bivariate association with mortality, but was not statistically significant when added to the model with either PSI or CURB‐65
		90‐day mortality	N/A	N/A	N/A	OR = 1.18 (95% CI: 1.05–1.32)	N/A
Yang et al. 19	67	In‐hospital mortality (in patient)	Median in patient who died (11.96; IQR: 7.26–30.68)	82.61	72.2	N/A	0.799	Age, sex, PCT, CRP, comorbidities, PSI, CURB‐65	NLR is a simple promising marker for predicting in‐hospital mortality
		In‐hospital mortality (ICU)	Median in patient who did not die (4.19; IQR : 2.39–7.52)	N/A	N/A	N/A	N/A
			Cutoff = 7.12
Kaya 20	N/A	In‐patient mortality	N/A	N/A	N/A	N/A	N/A	Age, sex, comorbidities, PSI, CURB, laboratory values	NLR can be used in estimating mortality, but is not superior to the commonly used scoring system (PSI, CURB‐65)
		ICU Mortality	Dead NLR = 13.5 ± 9	N/A	N/A	N/A	0.743 (95% CI: 0.627–0.860)
			Survived NLR = 7.9 ± 6.8	N/A	N/A	N/A

Abbreviations: AUC, area under the ROC curve; CI, confidence interval; CRP, C‐reactive protein; CURB‐65, Confusion, Respiratory rate, Blood pressure, 65 years of age and older; HR, high risk; HR, hazard ratio; ICU, intensive care unit; IQR, interquartile range; LR, low risk; MR, medium risk; N/A, not available; NLR, neutrophil–lymphocyte ratio; NT‐proBNP, N‐terminal (NT)‐prohormone B‐type natriuretic peptide; OR, odds ratio; PCT, procalcitonin; PSI, Pneumonia Severity Index; WBC, white blood cell.