Comparison of diagnostic accuracy in sepsis between presepsin, procalcitonin, and C-reactive protein: a systematic review and meta-analysis

Annals of Intensive Care

Table 4 Summary of subgroup analysis of the included studies by the potentially influencing variables

Variable	Category	Number of studies	Number of patients	Pooled sensitivity	Pooled specificity
Asian country	Yes	8	2091	0.81 (0.76–0.86)	0.82 (0.78–0.85)
Asian country	No	10	1379	0.86 (0.80–0.90)	0.70 (0.54–0.83)
Patients admitted in ICU	Yes	8	783	0.86 (0.80–0.90)	0.64 (0.51–0.76)
Patients admitted in ICU	No	10	2687	0.83 (0.76–0.89)	0.82 (0.73–0.88)
Whole blood specimen	Yes	6	1942	0.83 (0.73–0.88)	0.82 (0.78–0.86)
Whole blood specimen	No	12	1528	0.84 (0.79–0.89)	0.71 (0.58–0.82)
Sample sizes great than 150	Yes	6	2215	0.78 (0.72–0.83)	0.77 (0.68–0.84)
Sample sizes great than 150	No	12	1255	0.87 (0.82–0.90)	0.75 (0.62–0.84)
Proportions of patients with sepsis great than 0.5	Yes	14	2802	0.84 (0.80–0.88)	0.72 (0.63–0.79)
Proportions of patients with sepsis great than 0.5	No	4	668	0.85 (0.55–0.96)	0.86 (0.63–0.95)
Cut-off value great than 700	Yes	4	642	0.85 (0.70–0.93)	0.59 (0.37–0.78)
Cut-off value great than 700	No	14	2828	0.84 (0.79–0.87)	0.80 (0.72–0.85)