Systematic review regarding metabolic profiling for improved pathophysiological understanding of disease and outcome prediction in respiratory infections

Table 4 Summary of selected literature relevant to intermediate energy metabolism in all-cause mortality prediction

First author, year, reference	Marker	Study type	Study population	Key findings	Limitations
Ramakrishna et al., 2012, [107]	Lactate	Single-center, observational prospective cohort-study (follow-up until hospital discharge)	233 Malawian children with pneumonia	- The odds ratio for in-hospital mortality (25 deaths) in children with lactate >2 mmol/L was 7.48 (95 % CI 1.72–32.6) compared with children with lactate <2 mmol/L	- Single-center study
Ramakrishna et al., 2012, [107]	Lactate		233 Malawian children with pneumonia		- Low statistical power (only 25 deaths)
Chen et al., 2015, [108]	Lactate	Single-center, observational prospective cohort-study (28-day follow-up)	1,641 patients with pneumonia (861 inpatients, 780 outpatients) (Emergency Department of Beijing Chao-Yang Hospital)	- Non-survivors had higher lactate and CURB-65 scores compared with survivors (p < 0.001)	- Single-center study
					- No information about microorganisms
				- Lactate predicted 28-day mortality better than CURB-65 score (AUC 0.823 vs. 0.692; p < 0.01)
				- Combination of lactate and CURB-65 score improved the predictive value of CURB-65 score alone (AUC 0.851)

AUC area under the receiver operating characteristic curve, CI confidence interval, CURB-65 score new-onset confusion, urea >7 mmol/L, respiratory rate ≥30 breaths per minute, systolic or diastolic blood pressure <90 mmHg or ≤60 mmHg, respectively, age ≥65 years (pneumonia/LRTI risk scoring system), p p-value are statistically significant at p < 0.05

ISSN: 1465-993X