Selection And Characteristics Of Studies
As a result of the literature search, a total of 298 studies were identified, including 97 from PubMed, 91 from Embase, 62 from CNKI, and 48 from Wanfang debase. Figure 1 showed the study selection process. 111 duplicate publications were excluded. We excluded 145 studies for various reasons by evaluating the titles and abstracts. The remaining 42 studies were further scrutinized by reading the full text. Finally, only 19 studies were included in this meta-analysis, of which 9 reported the predictive value on disease severity [26–34], 6 reported the predictive value on mortality [35–40], and 4 reported the predictive value on both disease severity and mortality [41–44].
The characteristics of the included studies and the predictive value of NLR on disease severity or mortality in each study were presented in Table 1. Most studies were conducted in China. Twelve studies were published in English, six in Chinese and one in Spanish. Except one prospective study [27], all others were retrospective studies.
Table 1
Characteristics of the included studies and diagnostic test performance of NLR for disease severity and mortality
|
Study
|
Country
|
Publication language
|
No. of
patients
|
Male/
female
|
Mean age
|
Cut-off
|
AUC
|
TP
|
FP
|
FN
|
TN
|
SEN (%)
|
SPE (%)
|
Severity
|
Yang2020[33]
|
China
|
English
|
93
|
56/37
|
46.4 ± 17.6
|
3.3
|
0.84
|
21
|
25
|
3
|
44
|
88.0
|
63.6
|
|
Wang2020[31]
|
China
|
English
|
45
|
23/22
|
39.0 ± 11.5
|
13.4
|
0.89
|
8
|
6
|
2
|
29
|
83.3
|
82.4
|
|
Fesih2020[44]
|
Turkey
|
English
|
139
|
62/77
|
55.5 ± 18.5
|
3.3
|
0.87
|
43
|
25
|
11
|
60
|
79.0
|
71.0
|
|
Jingyuan Liu2020[27]
|
China
|
English
|
115
|
64/51
|
NA
|
3.1
|
NA
|
28
|
13
|
9
|
65
|
75.7
|
83.3
|
|
Asghar2020
|
Pakistan
|
English
|
100
|
69/31
|
52.6 ± 15.7
|
3.7
|
0.80
|
29
|
25
|
4
|
42
|
87.9
|
62.1
|
|
Sun2020[30]
|
China
|
English
|
116
|
60/56
|
50.0 ± 4.0
|
4.5
|
0.89
|
20
|
9
|
7
|
80
|
74.1
|
89.9
|
|
Shang2020[29]
|
China
|
English
|
443
|
220/223
|
56.0 ± 17.4
|
4.3
|
0.74
|
78
|
50
|
61
|
254
|
56.3
|
83.7
|
|
Yueping Liu2020[28]
|
China
|
English
|
84
|
47/37
|
53.0 ± 17.8
|
4.9
|
0.76
|
13
|
8
|
10
|
53
|
56.5
|
86.9
|
|
Basbus2020[42]
|
Spain
|
Spanish
|
131
|
71/60
|
52.0 ± 30.4
|
3.0
|
NA
|
17
|
36
|
4
|
74
|
80.9
|
67.3
|
|
Li2020[43]
|
China
|
Chinese
|
93
|
55/38
|
62.1 ± 16.8
|
11.3
|
NA
|
34
|
4
|
9
|
46
|
79.1
|
92.0
|
|
Zha2020[34]
|
China
|
Chinese
|
85
|
57/28
|
54.2 ± 16.0
|
5.6
|
0.77
|
25
|
10
|
12
|
38
|
68.8
|
78.4
|
|
Fei2020[26]
|
China
|
Chinese
|
72
|
32/40
|
58.0 ± 13.8
|
3.0
|
0.89
|
20
|
14
|
0
|
38
|
100
|
73.1
|
|
Xia2020[32]
|
China
|
Chinese
|
63
|
33/30
|
63.4 ± 14.9
|
4.8
|
0.83
|
26
|
8
|
5
|
24
|
83.9
|
75.0
|
Mortality
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cheng2020[36]
|
China
|
English
|
456
|
211/245
|
55.0 ± 18.6
|
3.2
|
0.81
|
26
|
107
|
10
|
303
|
78.3
|
73.9
|
|
Tatum2020[38]
|
America
|
English
|
125
|
57/68
|
58.7 ± 14.8
|
10.0
|
0.71
|
12
|
4
|
11
|
98
|
52.4
|
96.7
|
|
Chen2020[35]
|
China
|
English
|
681
|
362/219
|
65.0 ± 13.3
|
6.7
|
0.86
|
87
|
130
|
17
|
447
|
83.7
|
77.5
|
|
Fesih2020[44]
|
Turkey
|
English
|
139
|
62/77
|
55.5 ± 18.5
|
5.7
|
0.85
|
11
|
13
|
2
|
113
|
83.0
|
90.0
|
|
Asghar2020[41]
|
Pakistan
|
English
|
100
|
69/31
|
52.6 ± 15.7
|
4.2
|
0.81
|
20
|
29
|
2
|
49
|
90.9
|
62.6
|
|
Yan2020[39]
|
China
|
English
|
1004
|
493/511
|
NA
|
11.8
|
0.95
|
39
|
211
|
1
|
753
|
97.5
|
78.1
|
|
Basbus2020[42]
|
Spain
|
Spanish
|
131
|
71/60
|
52.0 ± 30.4
|
3
|
NA
|
7
|
46
|
2
|
76
|
77.8
|
62.3
|
|
Li2020[43]
|
China
|
Chinese
|
93
|
55/38
|
62.1 ± 16.8
|
11.3
|
0.92
|
28
|
10
|
3
|
52
|
90.3
|
83.9
|
|
Song2020[37]
|
China
|
Chinese
|
84
|
56/28
|
66.5 ± 12.2
|
6.1
|
0.87
|
32
|
5
|
10
|
37
|
76.2
|
88.1
|
|
Zhang2020[40]
|
China
|
Chinese
|
154
|
81/73
|
69.2 ± 7.5
|
9.4
|
0.86
|
21
|
10
|
6
|
117
|
76.2
|
92.0
|
AUC, area under curve; TP, true positive; FP, false positive; FN, false negative; TN, true negative; SEN, sensitivity; SPE, specificity; NLR, neutrophil-to-lymphocyte ratio; NA, not available. |
The number of participants across studies ranged from 45 to 1004. Notably, the SEN, SPE, AUC and cut-off value of NLR predicting mortality and disease severity ranged greatly among the included studies. Except two studies [41, 42], all other studies defined severe patients as meeting at least one of the following criterions: Shortness of breath, respiratory rate (RR) ≥ 30 times/min, or oxygen saturation (resting state) ≤ 93%, or PaO2/FiO2 ≤ 300 mmHg.
Study Quality And Publication Bias
The methodological quality of the included studies was presented in Additional File 2. One study only included patients classified as moderate [36], one only included severe patients [35], and one only included elderly patients [40]. Therefore, these three studies were considered to have a high risk of patient selection. Four studies were also assessed to be with high risk of patient selection, because they did not avoid a case-control design [32, 34, 37, 43]. One study did not provide sufficient information about patients enrolled, and we considered this leaded to a high risk of patient selection [33].Most studies were considered to have unclear risk of bias regarding index tests, because they did not report the blindness between index and reference tests. Deek funnel plot was shown in Additional File 3, publication bias may exist among studies reporting the predictive value of NLR on disease severity (P = 0.04).
Predictive Value Of NLR On Disease Severity
Thirteen studies involving 1579 patients reported the predictive value of NLR on disease severity. The pooled SEN and SPE were 0.78 (95% CI 0.70–0.84, I²=71.83) and 0.78 (95% CI 0.73–0.83, I²=77.80), respectively (Fig. 2a). The positive likelihood ratio was 3.6 (95% CI 2.9–4.4), and the negative likelihood ratio was 0.28(95% CI 0.21–0.38). The DOR was 13(95% CI 9–18). The SROC curve was shown in Fig. 3a. The AUC of NLR for predicting disease severity was 0.85 (95% CI 0.81–0.88), indicating high diagnostic value. We can learn from Fagan nomogram (Fig. 4a) that if the pre-test probability was set to 50%, the post-test probability of NLR for the detection of severe cases was 78% when the NLR was above the cut-off valve. On the contrary, when the NLR was below the cut-off valve, the post-test probability was 26%.
Predictive Value Of NLR On Mortality
Ten studies involving 2967 patients reported the predictive value of NLR on mortality. The pooled SEN and SPE were 0.83 (95% CI 0.75–0.89, I²=66.13) and 0.83 (95% CI 0.74–0.89, I²=90.34), respectively (Fig. 2b). The positive likelihood ratio was 4.8 (95% CI 3.3-7.0), and the negative likelihood ratio was 0.21(95% CI 0.15–0.30). The DOR was 23(95% CI 15–36). The SROC with pooled diagnostic accuracy was 0.90 (95% CI 0.87–0.92), presented in Fig. 3b. The Fagan nomogram showed that the post-test probability of NLR for the detection of mortality was 83% when the NLR was above the cut-off valve and the post-test probability was 17% when the NLR was below the cut-off valve. (Fig. 4b).
Subgroup Analyses And Sensitivity Analyses
We conducted the subgroup analyses based on the cut-off value. In terms of predicting disease severity, the cut-off value in six studies was higher than 4.5, and were termed the “high cut-off value” subgroup. Seven others used a lower cut-off value, which were included in the “low cut-off value” subgroup. The AUC were 0.86 (95% CI 0.83–0.89) and 0.82 (95% CI 0.78–0.85), respectively. Similarly, ten studies reporting the predictive value of NLR on mortality were divided into “high cut-off value” (cut-off ≥ 6.5) and “low cut-off value”(< 6.5) subgroups, and the AUC were 0.92 (95% CI 0.89–0.94) and 0.84 (95% CI 0.80–0.87), respectively. In the sensitivity analyses, we only included studies published in English. The pooled AUC for predicting disease severity and mortality were 0.83 (95% CI 0.80–0.86) and 0.90 (95% CI 0.87–0.92), respectively. Detailed results about subgroup analyses and sensitivity analyses were presented in Table 2.
Table 2
Results of sensitivity analysis and subgroup analysis
Categories
|
No. of
studies
|
Sensitivity
(95% CI)/I²
|
Specificity
(95% CI) /I²
|
AUC
(95% CI)
|
DOR
(95% CI)
|
PLR/NLR
|
Disease severity
|
|
|
|
|
|
|
Cut-of ≥ 4.5
|
6
|
0.74(0.66,0.80)/25.56
|
0.86(0.81,0.89)/36.40
|
0.86(0.83,0.89)
|
17(10,28)
|
5.1/0.31
|
Cut-of < 4.5
|
7
|
0.82(0.71,0.89)/82.74
|
0.72(0.66,0.78)/79.70
|
0.82(0.78,0.85)
|
12(7,19)
|
3.0/0.25
|
Published in English
|
8
|
0.74(0.63,0.83)/73.82
|
0.78(0.71,0.84)/81.99
|
0.83(0.80,0.86)
|
10(7,16)
|
3.4/0.33
|
Mortality
|
|
|
|
|
|
|
Cut-off ≥ 6.5
|
5
|
0.83(0.66,0.92)/84.97
|
0.87(0.77,0.93)/92.60
|
0.92(0.89,0.94)
|
32(17,61)
|
6.3/0.20
|
Cut-off < 6.5
|
5
|
0.81(0.72,0.87)/0
|
0.77(0.64,0.86)/89.07
|
0.84(0.80,0.87)
|
14(7,27)
|
3.5/0.25
|
Published in English
|
6
|
0.83(0.69,0.91)/79.98
|
0.82(0.71,0.90)/89.87
|
0.90(0.87,0.92)
|
23(12,41)
|
4.7/0.21
|
AUC, area under curve; PLR, positive likelihood ratio; NLR, negative likelihood ratio; DOR, diagnostic odds ratio; CI, credible interval. |