Imagine this clinical scenario: you are presented with a dog whose clinical and radiographic findings are consistent with a diagnosis of gastric dilatation volvulus (GDV)/gastric torsion. You advise initial stabilisation followed by surgical de-rotation and gastropexy as treatment.
Before embarking on surgery with high associated costs, the owner would like to know the likelihood of the dog’s survival to discharge. You are aware that a measurement of blood lactate concentration may help answer this question. So, you decide to consult the evidence regarding admission or pre-operative blood lactate level as a predictor of survival to discharge.
The evidence
Fifteen studies (12 retrospective and three prospective) were identified and appraised based on the predefined inclusion criteria. Four further studies examining lactate concentration in dogs with a range of conditions (not restricted to GDV) and two review papers were also appraised.
Six studies had the PICO question as their primary study focus: Rauserova-Lexmaulova et al. (2020), Grassato et al. (2020), Santoro Beer et al. (2013), Green et al. (2011), Zacher et al. (2010) and de Papp et al. (1999).
Most of the studies did not have the PICO question as their primary objective, which meant that in some cases, statistical analysis included less detail or focus on the PICO question.
Of the 15 studies, 12 were retrospective, with the limitations inherent to retrospective studies. In particular, there was variation in the presenting signs, characteristics of the study population and analysis and presentation of data in and/or between studies.
The median overall survival rate after surgery across all studies was 78.4 percent, while the range was 64.1 to 88 percent.
Pre-intervention plasma lactate measurements
All 15 studies reported an initial (pre-intervention) lactate level in survivors versus non-survivors using a variety of reporting metrics.
All 15 studies reported an initial (pre-intervention) lactate level in survivors versus non-survivors using a variety of reporting metrics
Eleven papers reported a significant difference in initial lactate concentration between survivors and non-survivors, while four studies reported no significant difference in initial lactate concentration between the groups.
Receiver operating characteristic analyses
Six studies performed receiver operating characteristic analyses to determine a cut-off for lactate which optimised sensitivity and specificity for survival versus non-survival. However, the selected optimal cut-offs across studies varied between a lactate of 4.1 and 9.0 mmol/l, which makes this data difficult for clinicians to use when making decisions regarding individual patients.
In all studies, the sensitivity was moderately low (range 60.3 to 75.4 percent), while the specificity was somewhat higher (range 73 to 90.9 percent).
The selected optimal cut-offs across studies varied between a lactate of 4.1 and 9.0 mmol/l, which makes this data difficult for clinicians to use
Change in lactate concentration following intervention (lactate clearance)
Five papers examined changes in lactate concentration (lactate clearance) over a determined period of time and its relationship to survival.
Of these, four papers (Zacher et al., 2010; Green et al., 2011; White et al., 2021; Verschoof et al., 2017) reported lactate concentration reduced to a greater extent in survivors compared to non-survivors (as well as showing a difference in static initial lactate concentration). It is difficult to compare these studies directly, as the time points for subsequent lactate measurements differed.
Grassato et al. (2020) was the only study not to find a difference in initial lactate in survivors versus non-survivors, and this study had several limitations.
Zacher et al. (2010) looked at lactate pre- and post-resuscitation prior to surgery and showed a more significant reduction in lactate in survivors compared to non-survivors. Using a cut-off of change in lactate of 42.2 percent predicted survival with a sensitivity of 61 percent and specificity of 100 percent. This suggests that a dog displaying a reduction in lactate of 42.2 percent should be given a good prognosis, whereas failing to show this reduction in lactate should be interpreted more cautiously.
Non-GDV studies looking at lactate as a prognostic marker
Several studies examined lactate as a prognostic marker in cases other than dogs with GDV.
Stevenson et al. (2007) (n=80) and Blutinger et al. (2021) (n=71) both found no significant difference in initial lactate levels between survivors and non-survivors. Additionally, both found that a reduction in lactate post-resuscitation was a significant predictor of survival.
In contrast, Kohen et al. (2018) (n=566) and Ortolani and Bellis (2021) (n=267) found a significant difference in admission lactate concentration between survivors and non-survivors. Kohen et al. (2018) additionally found odds of mortality increased with an increasing magnitude of hyperlactataemia. However, both studies also reported a significant overlap in lactate values between survivors and non-survivors.
It should be noted that the studies with smaller populations were more likely to be underpowered to detect a significant difference in lactate concentration between survivors and non-survivors.
Analysis
In all studies, the range of lactate values reported for survivors and non-survivors was very wide. Further, there was significant overlap in values between the two groups in all cases.
Many studies had small sample sizes; therefore, those that reported non-significance could have been underpowered to detect significance in plasma lactate levels between survivors and non-survivors (no power analysis was undertaken). Even in studies where a significant difference was found between populations, the broad range of values and substantial overlap between survivors and non-survivors makes prediction at an individual case level difficult. This is true even though it may be correctly inferred that a high lactate value is more commonly associated with non-survival compared to a lower lactate level.
The problem with low sensitivity is the relatively high false negative rate. Approximately 25 percent of dogs with a sensitivity of 75 percent would be falsely classified as having low lactate and, therefore, may be given a falsely optimistic prognosis. This may affect the decision to proceed to surgery, with cost implications for the owner, but lead to a poor outcome.
In general, a “low” plasma lactate concentration is a better predictor of survival than a “high” plasma lactate concentration is a predictor of non-survival
Conversely, the false positive rate corresponding to a low specificity means dogs may be given a falsely poor prognosis, ie they may have a good outcome, but their owners may be advised not to proceed with surgery and consider euthanasia instead. However, in general, a “low” plasma lactate concentration is a better predictor of survival than a “high” plasma lactate concentration is a predictor of non-survival.
The papers that examined changes in lactate concentration over a determined period and its relationship to survival showed that, at population level, dogs that reduce their lactate concentration (compared to admission level) are likely to have a better prognosis than those dogs who do not. However, it is difficult to use this information to make decisions for individual animals.
Conclusion
Overall, the findings suggest that, in general, a dog with lower lactate, or which shows a reduction in lactate following fluid resuscitation, should be given a better prognosis, and corrective surgery should be recommended. A dog with a higher lactate, or which does not decrease following adequate fluid resuscitation, should cautiously be given a more guarded prognosis.
Owners should be advised that many dogs with higher lactate, and those that do not have a decrease in lactate following fluid resuscitation, may still have a reasonable expectation of survival
However, owners should be advised that many dogs with higher lactate, and those that do not have a decrease in lactate following fluid resuscitation, may still have a reasonable expectation of survival, and exploratory laparotomy should still be strongly considered.
The full Knowledge Summary can be read in RCVS Knowledge’s open access journal Veterinary Evidence. |
Disclaimer
The application of evidence into practice should take into account multiple factors, not limited to individual clinical expertise, patient’s circumstances, owner’s values, country, location and/or clinic where you work, the individual case in front of you and the availability of therapies and resources.
Knowledge Summaries are a resource to help reinforce or inform decision making. They do not override the responsibility or judgement of the practitioner to do what is best for the animal in their care.