Skip to main content

Table 3 Studies with comparison of arterial, venous, and capillary lactate.

From: Blood lactate as a predictor for in-hospital mortality in patients admitted acutely to hospital: a systematic review

Author, year, reference

Study design, study quality*

N

Patient characteristics

Lactate characteristics

Outcome measure

Cut off [mM]

Main findings

Lavery et al., 2000 [26]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

231

Trauma patients, unselected.

Arterial and peripheral venous lactate within 10 min after admission at trauma center.

Comparison of arterial and venous lactate.

2.0

No significant difference between arterial and venous lactate was found. The correlations were: a.femoralis-v.femoralis, r = 0.95; a.femoralis-peripheral vein, r = 0.935; a.radialis-peripheral vein, r = 0.988. No significant difference was found for peripheral venous lactate drawn with or without the use of tourniquet. For LOS > 2 days, lactate > 2.0 mM had sens. 68%, spec. 41%, PPV 71%, and NPV 48%. For in-hospital mortality, lactate > 2.0 mM had sens. 95%, spec. 43%, PPV 16%, and NPV 99%.

Boldt et al., 2001 [34]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

40

Surgical intensive care patients, unselected.

Arterial and capillary lactate at inclusion day, after 8 hours, and once per day the following three days.

Compares POCT and laboratory lactate. Comparison of arterial and capillary lactate.

2.0

Correlation between POCT and the hospital laboratory for measuring lactate in arterial blood was r2 = 0.97. Bland-Altman showed bias 0.15 mM with POCT tending to be lower. Correlation between capillary and arterial blood was r2 = 0.80. Bland-Altman showed bias 0.59 mM with capillary lactate tending to be higher.

Perez et al., 2008 [35]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

120

Patients in antiretroviral therapy with hyperlactatemia.

Venous lactate.

Compares POCT and reference-instrument.

2.2

Correlation between mean venous lactate measured with POCT and reference instrument (2.89 mM vs. 2.78 mM), r = 0.63. The POCT instrument had a sensitivity of 95.9%, and a specificity of 63.8% for hyperlactatemia. PPV was 80.5%. NPV was 90.9%. Agreement was analysed by Bland-Altman plot. Bias was 0.113 mM (-2.103-2.329) with POCT tending to show higher values.

Shapiro et al., 2010 [36]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

699

Patients with suspected infection (pneumonia etc.).

(≥ 18 years).

Venous lactate in the emergency department. Convenience sample.

Comparison of POCT venous lactate and laboratory venous control. In-hospital mortality.

 

Correlation between venous lactate measured by POCT and laboratory was 0.97. POCT lactate had area under ROC curve = 0.72 for mortality. Bland-Altman plot showed that POCT lactate was, on average, 0.32 mM (-0.35-0.98) lower than laboratory lactate, with agreement kappa = 0.97.

Gallagher et al., 1997 [37]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

69

Emergency department patients (≥ 18 years).

Arterial and peripheral venous lactate. Convenience sample.

Comparison of arterial and peripheral venous lactate.

1.6

Correlation between arterial and peripheral venous lactate was r2 = 0.89. There was no correlation between the arterio-venous difference and tourniquet time (r2 = 0.02). Peripheral venous lactate had a sensitivity of 94%, and a specificity of 57% for hyperlactatemia. Bias was 0.22 mM (0.04-0.41) with venous lactate tending to be higher.

Younger et al., 1996 [38]

Prospective, observational

Moderate

Acceptable external validity. Acceptable internal validity.

Sample size of uncertain adequacy.

48

Emergency department adult patients. (≥ 17 years).

Arterial and peripheral venous lactate.

Comparison of arterial and peripheral venous lactate.

1.6

Peripheral venous lactate > 1.6 mM has sensitivity of 100%, and specificity of 86% for hyperlactatemia. The correlation between arterial and venous lactate was r2 = 0.71 (p < 0.001). Bland-Altman plot showed bias 0.18 mM (0.012-0.372) with venous lactate tending to be higher.

  1. Numbers in brackets are 95% confidence interval unless specified otherwise. *quality rated by using methods validated for internal validity, precision, and applicability (external validity) [9]. LOS = length of stay; sens. = sensitivity; spec. = specificity; PPV = positive predictive value; NPV = negative predictive value; POCT = point-of-care testing; ROC = receiver operating characteristic.