• Dogs with gastric dilatation volvulus (GDV) confirmed radiographically or at necropsy.
• Mixed population represented.
• Lactate measured using Stat Profile Plus 9, Nova Biomedical.

• Correlation between initial admission plasma lactate concentration and survival to discharge.
• Correlation between initial admission plasma lactate concentration and gastric necrosis.

• Overall survival of dogs undergoing surgery was 87/101 (86%).
• Survival of dogs with lactate <6.0 mmol/L (69/70, [99%]) was significantly higher than dogs with lactate >6.0 mmol/L [18/31, 58%]) (sensitivity = 68%, specificity 88%; p <0.001).
• Logistic regression analysis showed an increasing probability of survival with decreasing pretreatment lactate level.

• Retrospective study.
• 23 dogs received fluid therapy prior to lactate measurement but these were included in the overall analysis together with dogs that did not receive such treatment.

• Dogs with gastric dilatation volvulus (GDV) confirmed radiographically and at surgery, no pretreatment prior to referral, > 1 measurement of plasma lactate level prior to surgery (pre and post-fluid therapy and stabilisation prior to surgery).
• Mixed population represented.
• Lactate measured using Stat Profile 9, Nova Biomedical.

• Overall survival 49/64 (77%).
• Initial plasma lactate for survivors (6.2 ± 3.2 mmol/L) was significantly lower than non-survivors (10.3 ± 3.2 mmol/L) p <0.05.
• Final lactate concentration (post fluids and decompression) for survivors (3.3 ± 2.3 mmol/L) was significantly lower than non-survivors (8.0 ± 3.3 mmol/L) p <0.05.
• Percentage change in lactate between initial level and the level post fluids and decompression was significantly different between survivors (49.1 ± 28.8%) and non-survivors (24.6 ± 19.4%) p <0.05, although absolute change was not significant.
• For all dogs, a cut-off for initial lactate of 9.0 mmol/L predicted survival with sensitivity 74% and specificity 73%.
• A cut-off for final lactate of 5.6 mmol/L predicted survival with sensitivity 84% and specificity 80%.
• A percentage change in lactate cut-off of 42.2% predicted survival with 61% and specificity 100%.
• Subset of dogs with 'high lactate' (HIL) were analysed separately. In this group, initial lactate did not differ between survivors and non-survivors.
• Final lactate (post fluids and decompression) was significantly different between survivors (5.1 ± 2.2 mmol/L) and non-survivors (9.9 ± 2.7 mmol/L) p <0.05.
• Absolute change in lactate was significantly different between survivors (6.3 ± 2.0 mmol/L) and non-survivors (2.6 ± 2.3 mmol/L) p <0.05.
• Percentage change in lactate was significantly different between survivors (53.2 ± 17.1%) and non-survivors (18.9 ± 20.5%) p <0.05.
• For this subset of dogs, a cut-off of 6.4 mmol/L in final lactate (post fluid and decompression) predicted survival with sensitivity 77% and specificity 91%.
• For this subset of dogs, a lactate cut-off of 4.0 mmol/L in absolute change in lactate predicted survival with sensitivity 92% and specificity 82%.
• For this subset of dogs, a cut-off of 42.5% in percentage change in lactate predicted survival with sensitivity 85% and specificity 100%.

• Retrospective study.
• Small sample size (power analysis not reported).
• Three dogs with initial plasma lactate concentration within the reference range were excluded from the study.
• Variation in pre-surgery fluid therapy.
• No definitive endpoint for resuscitation was reported.
• Non-survivors included dogs euthanised at surgery based on subjective impression of gastric necrosis, which may have introduced bias.

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV), admission blood lactate level measured and subsequent surgery.
• Mixed population represented.
• Lactate measured using Stat Profile 9, Nova Biomedical.

• Correlation between admission blood lactate level and survival to discharge.
• Correlation between change in plasma lactate levels (lactate clearance) and survival to discharge.
• Correlation between admission blood lactate level and presence of gastric wall necrosis.
• Correlation between gastric wall necrosis and survival to discharge.

• Overall survival 74/84 (88%).
• Significant difference in admission lactate concentration between non-survivors (median 6.80 mmol/L [range 1.4–16.9 mmol/L]) and survivors (median 3.4 mmol/L [range 0.7–16.1 mmol/L]), p <0.0074.
• Of 74/84 (88%) that survived, 55/74 (74.4%) of dogs had initial blood lactate <6.0 mmol/L, while 19/74 (25.6%) of dogs had initial blood lactate ≥0 mmol/L.
• Initial blood lactate <4.1 mmol/L had a sensitivity of 60.3% and specificity 90.9% for predicting survival (P = 0.0133).
• Serial lactate measurements were available for 52/84 (62%) of dogs. Of these 52 dogs, 40/52 (77%) had a raised (>2.5 mmol/L) initial lactate concentration, and 37/40 (92.5%) of these dogs survived. 26/37 (70%) of these survivors had a reduction in serial lactate of >= 50% from baseline. Of the3/40 (7.5%) of dogs that did not survive, the serial lactate failed to fall ≥ 50% from baseline.

• Retrospective study.
• Some dogs had treatment prior to referral and were not excluded, which may have introduced bias.
• According to the protocol described in the paper, the treatment that each dog received may have been variable (e.g., some dogs may have received transfusion).
• Serial lactate results were difficult to interpret, with variability in timing of serial lactate measurements. Also, not clear if some serial lactate results were pre-surgery or post-surgery. Statistical analysis was also not performed on this, including no multivariate analysis, and this may be because the group size was too small. The statistics presented do not readily support the assertion that reduction in lactate from baseline may be a better prognostic indicator than single baseline lactate measurement.

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV) with measurement of blood lactate level at presentation.
• Mixed population represented.
• Lactate measurement method not specified.

• Dogs underwent corrective surgery for GDV.
• Other resuscitative interventions not specified.

• Survival to discharge correlated with plasma lactate level.
• Correlation between survival and risk factors for survival (signalment, time of presentation, presenting clinical signs), decompression, and thoracic radiographic findings.

• Overall survival rate 85/101 (84%).
• Evidence for cardiomegaly on preoperative thoracic radiographs was associated with decreased survival.
• After controlling for cardiomegaly, lactate <6 mmol/L at presentation had an odds ratio of 7.3 of survival compared with dogs with lactate ≥ 6.

• Retrospective study.
• Lactate cut-off results were reported after controlling for the presence of cardiomegaly.
• Case selection made on basis of presence of thoracic radiographs which was incomplete, which may have introduced bias.
• Not clear from the data why a lactate cut-off of 6 mmol/L was taken for calculating odds ratio.
• Possibilities of Type I and Type II errors due to incomplete records.

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV), confirmed at surgery, and admission blood lactate level measured.
• Mixed population represented.
• Lactate measured using either Advia 120 (Siemens Medical) or Abacus or Arcus (Diatron).

• Correlation between plasma lactate concentration at admission and survival to discharge.
• Correlation between plasma lactate concentration at admission and degree of gastric wall necrosis.
• Correlation between plasma markers serum canine pepsinogen-A, CRP, and cPLI and survival to discharge and degree of gastric wall necrosis.

• Overall survival 51/66 (77.3%).
• No significant difference in lactate concentration between non-survivors (7.9 mmol/L [range 1.1–25.3 mmol/L] and survivors (4.8 mmol/L [range 0.1–19.1 mmol/L]).

• Retrospective study.
• Non-survivors included dogs euthanised for a variety of reasons including subjective impression of gastric necrosis, owner request, or deterioration during hospitalisation.
• Variation in resuscitative intervention between dogs not specified.
• Lactate concentration was analysed using two different analysers.
• Power analysis not reported.

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV) with no concurrent disease with no prior intervention, and venous blood sample at time of admission, and subsequently underwent corrective surgery for GDV.
• Mixed population represented.
• Lactate measured using Stat Profile Critical Care Xpress, Nova Biomedical.

• Correlation between admission blood lactate level and survival to discharge.
• Correlation between presence or absence of gastric necrosis and survival to discharge.
• Correlation between admission base excess and presence of gastric necrosis and survival to discharge.

• Overall survival rate 65/78 (83%).
• Initial plasma lactate was significantly associated with survival (non-survivors 9 mmol/L [range 5.6–15 mmol/L), survivors 4.5 mmol/L [range 0.8–14.54 mmol/L] P <0.001).
• A cut-off admission lactate value of 7.4 mmol/L had sensitivity 75% and specificity 89% for predicting survival to discharge.

• Retrospective study.
• Potential variation in treatment based on initial plasma lactate may have introduced bias.
• Study included patients euthanised for financial or other unknown reasons.

• Dogs presenting alive or dead with a confirmed or presumptive diagnosis of gastric dilatation volvulus (GDV), with lactate measured for cases presenting alive.
• Mixed population represented.
• Lactate measurement method not specified.

181/492 cases (36.8%) had blood lactate levels measured.

• Survival to discharge vs non-survival correlated with admission (pretreatment) blood lactate level.
• Population regression modelling to identify risk factors associated with presumptive diagnosis of GDV.

• Overall survival (of dogs undergoing surgery) was 226/285 (79.3%).
• For dogs arriving alive, blood lactate concentration at presentation was not associated with likelihood of surgery.
• Dogs with blood lactate concentrations <4 mmol/L had an increased probability of survival to discharge both overall (P <0·001) and among the surgical cases (P <0·001).

• Retrospective study.
• Multi-centre study with variability in case decision-making and surgery.
• Diagnosis of GDV was presumptive in some cases.
• Lactate concentration was not complete (36.8% of presenting population) therefore bias relating to missing data may have been present.
• Missing data in case records may not have been a random effect which may have introduced bias.
• 198/492 (40.2%) of cases did not undergo surgery (of which 184/198 [92.9%]) were euthanised. Also, of the 285/492 (57.9%) of dogs that did undergo surgery, 59/285 (20.7%) did not survive, of which 37.59 (62.7%) were euthanised. As the decision to euthanise involves a number of factors, this may have introduced bias into the overall survival rate.

• Dogs with gastric dilatation volvulus (GDV) confirmed radiographically and for which blood samples were available.
• Mixed population represented.
• Lactate concentration measured using blood gas analyser cobab b 221 (Roche Diagnostics, Switzerland).

• All dogs received preoperative stabilisation with intravenous fluid therapy and gastric needle decompression.
• All dogs subsequently underwent corrective surgery for GDV.

• Correlation between survival to discharge and initial blood lactate level.
• Correlation between initial blood levels of coagulation variables (platelet count, prothrombin time, activated partial thromboplastin time, fibrinogen, antithrombin, protein C, protein S, and D-dimers) and inflammatory biomarkers (CRP, WBC count, lymphocyte and neutrophil numbers) and survival to discharge.

• Overall survival 15/20 (65%).
• Initial plasma lactate was significantly higher in non-survivors (11.8mmol/L [range 7.5–16.2 mmol/L] than overall survivors (6.2mmol/L [range 1.9–9.7 mmol/L]) p <0.01.
• Plasma lactate on day 1 post surgery was significantly different between non-survivors and survivors (p = 0.0078). (Values not given, but significant overlap between populations displayed graphically).
• For surviving dogs, plasma lactate decreased significantly between initial lactate, day 1 and day 2 post-surgery (this data was not compared with non-survivors).

• Small sample size (power analysis not reported).
• Two dogs were euthanised during surgery due to subjective impression of severe gastric necrosis.
• Five dogs were euthanised post-surgery due to sepsis and disseminated intravascular coagulation (DIC).

• Dogs with gastric dilatation volvulus (GDV) that underwent corrective surgery and which had blood samples taken from both cephalic and saphenous veins.
• Mixed population represented.
• Lactate measured using a 'dry' analyser or 'wet' chemistry autoanalyser (Reflotron®, Boehringer Mannheim or Reflovet Plus, Roche).

• Fluid resuscitation and gastric decompression (details varied between subjects and between centres).
• All dogs underwent corrective surgery.

• Survival to discharge vs non-survival correlated with admission (pretreatment) blood lactate concentration.
• Comparison of lactate concentration between cephalic and saphenous sampling sites.

• Overall survival 37/45 (82.9%).
• No significant difference detected in lactate between the saphenous and cephalic sampling sites.
• Initial median saphenous and cephalic lactate concentration for all dogs was significantly higher in non-survivors (3.9 mmol/L [range 0.7–17.8 mmol/L] and 3.8 mmol/L [range 0.78–21.9 mmol/L] from saphenous and cephalic sites respectively), compared to survivors (11.2 mmol/L [range 6.6–11.8 mmol/L] and 11.4 mmol/L [range 3.8–19.4 mmol/L] respectively), p = 0.01.
• When dogs that had undergone gastric decompression prior to blood sampling (n = 10) were removed from the above analysis results remained significant at: non-survivors 10.7mmol/L [range 6.8–11.2 mmol/L] and 11.1mmol/L [range 3.8–11.4 mmol/L] and survivors 3.5 mmol/L [range 0.7–17.8 mmol/L] and 3.1mmol/L [range 0.7–19 mmol/L] (from saphenous and cephalic respectively) p <0.001.

• Dogs that died or were euthanised before surgery were excluded from analysis which potentially biases the results by restricting analysis to a more stable population. The total study population was drawn from different populations, a) from a teaching hospital (n = 39) and b) from three specialty referral clinics (n = 6), these populations were combined in the subsequent analysis despite potential variation in resuscitation protocols and treatment.
• Lactate was analysed using different analysers.
• Reporting of results for all dogs included those that had already undergone an intervention (gastric decompression).

• Dogs diagnosed with gastric dilatation volvulus (GDV) and dogs diagnosed with intestinal obstruction (IO) (Control group).
• Mixed population represented.
• Lactate measured with portable lactate analyser (Lactate Scout +, EKF Diagnostics, Cardiff, UK).

• Survival to discharge vs non-survival correlated with admission (pretreatment) blood lactate level.
• Correlation between initial admission serum lipase activity, canine pancreatic lipase immunoreactivity (cPLI) and C-reactive protein concentration and survival.

• Overall survival 16/22 (73%).
• No significant difference between lactate level of non-survivors (7.1 mmol/L [range 3.1–12.1 mmol/L]) and survivors (5.25 mmol/L [range 1.8–10.3 mmol/L]).

• Basis for diagnosis of GDV not specified.
• Control group used was dogs with IO which is not a 'disease free' control group.
• Presurgical interventions mentioned but not specified and not clear if applied to all dogs.
• Some dogs (number not stated) were euthanised based on subjective impression of gastric wall necrosis, with potential for bias.
• Small number of dogs in study and authors state that number of dogs studied was lower than required to make assessment of mortality risk.

• Dogs with a clinical and radiographic diagnosis of gastric dilatation volvulus (GDV).
• Mixed population represented.
• Lactate measurement technique not specified.

• Preoperative stabilisation (details not specified).
• Corrective surgery for GDV.

• Survival to discharge vs non-survival correlated with admission (pretreatment) blood lactate level.
• Correlation between admission (pretreatment) levels of the biomarkers cell-free DNA, high mobility group box-1, procalcitonin, lactate, and APPLE_fast score, with survival, evidence of gastric necrosis, and occurrence of postoperative complications.

• Three dogs received treatment prior to presentation.
• Small sample and no power analysis performed.
• Four dogs were euthanised based on subjective assessment of gastric necrosis which may have biased the outcome results.
• Although blood samples were taken prospectively, records were analysed retrospectively with missing data or variation between subjects (e.g., timing between onset of clinical signs and admission / surgery).
• Two different centres contributed to the study with potential for differences in case management which may have affected outcomes.

• Dogs with a radiographic or surgical diagnosis of gastric dilatation volvulus (GDV) with no prior treatment, that subsequently underwent corrective surgery, and for which blood lactate measurement was available.
• Mixed population represented.
• Lactate measured on presentation (T0) and 24 hours (T24) and 48 hours (T48) after surgery.
• Method of lactate measurement not specified.

• All dogs received intravenous fluid resuscitation, antibiotic therapy and gastric decompression via orogastric tube and / or percutaneous needle.
• All dogs subsequently underwent corrective surgery for GDV.

• Survival to discharge vs non-survival correlated with T0, T24 and T48 blood lactate concentration and change in lactate concentration between time points.
• Presence or absence of gastric necrosis correlated with T0, T24 and T48 blood lactate concentration, and change in lactate concentration between time points.

• Overall survival 31/45 (69%).
• No significant difference detected in lactate concentration between at any of the three time points (all values in mmol/L).
• T0: Non-S 3.25 mmol/L (2.1–12.7 mmol/L), S 3.2 mmol/L (0.9–16.4 mmol/L); T24: Non-S 1.1 mmol/L (0.7–9.1 mmol/L), S 1.2 mmol/L (0.6–4.2 mmol/L); T48: Non-S NA (0.6–6.2 mmol/L), S 0.9 mmol/L (0.6–2.5 mmol/L). No significant difference in lactate concentration between either group (survivors versus non-survivors) at any time point.
• Despite the above, a lactate level of < 4.45 mmol/L was reported as distinguishing survivors from non-survivors with sensitivity 33% and specificity 77%.
• Lactate concentration for survivors (when reported as mean ± SD in table format) at T24 and T48 was significantly lower than at T0 whereas this was not the case for non-survivors (significance not given).

• Retrospective study.
• Method of lactate measurement not reported, nor how accurate timing of postoperative blood lactate levels was achieved.
• Small number of cases (power analysis not reported).
• Data regarding pre-hospitalisation factors was not available.
• The tabulated data showed survivors had a significant reduction in lactate between time-points compared to non-survivors when data was presented as mean ±SD, but not when reported as median with range. However, reductions were also reported in lactate levels in dogs both with and without gastric necrosis when presented as mean ±SD but not when reported as median with range. This led to a confusing discussion where the authors mixed discussion of gastric necrosis and outcome and concluded that reduction in lactate levels could not be used as a prognostic indicator.
• Apparent contradiction between results section and table as described above.
• Significance levels not reported – as described above.

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV) with no prior treatment, that subsequently underwent corrective surgery, with blood lactate samples taken at admission prior to any treatment.
• Mixed population represented.
• Initial (pretreatment) lactate measured only.
• Lactate measured using Konelab 20i analyser (Thermo Fisher Scientific, Finland) processed within 15 minutes or stored frozen and analysed the following day with same analyser.

• All subjects underwent corrective surgery for GDV.
• Other interventions not specified.

• Survival to discharge vs non-survival correlated with admission (pretreatment) blood lactate level.
• Presence or absence of gastric necrosis correlated with admission (pretreatment) blood lactate level.
• Correlation between initial values of pH, PCO₂, bicarbonate, base excess (BE) and electrolytes (sodium, potassium, chloride, total calcium and inorganic phosphorous), anion gap (AG), strong ion difference (SID) and surgical finding of gastric necrosis and survival to discharge.

• Overall survival 61/75 (81.3%.)
• Initial plasma lactate concentration was significantly associated with mortality (non-survivors median 9.68 mmol/L, [range 5.84–13.49 mmol/L] vs. survivors, median 3.36 mmol/L, [range 2.42–5.64 mmol/L] p ˂ 0.0001).
• However, in multivariate logistic regression analysis, lactate concentration was not independently associated with mortality.
• Optimal cut-off points to predict patient survival were lactate ≤ 5.62 mmol/L (75.4% sensitivity, 85.7% specificity).

• Retrospective study.
• Samples from both arterial and venous sources were used for the comparison of lactate.
• Pre-surgical intervention / stabilisation was not specified.
• Risk of possible concurrent diseases (e.g., cardiac, liver, and renal) could not be excluded.
• Relatively small number of patients (power analysis not undertaken).

• Dogs with a radiographic diagnosis of gastric dilatation volvulus (GDV) with no prior treatment.
• Mixed population represented.
• Method of lactate measurement not specified.

• Survival to discharge vs non-survival correlated with presentation lactate measurement.
• Non-survivors were further categorised as: euthanised preoperatively, euthanised intraoperatively, euthanised postoperatively, and died.

• Overall survival 275/429 (64.1%).
• Initial plasma lactate (3.7 mmol/L [range 0.5–20 mmol/L]) was significantly different between survivors and those that were euthanised preoperatively (6.1 mmol/L [range 1.5–19.4 mmol/L]), intra-operatively (8.15 mmol/L [range 2.7–15.7 mmol/L]), and died postoperatively (10.1 mmol/L [1.6–19 mmol/L]) (p <0.001 for all), but not for those that were euthanised postoperatively (5.65 mmol/L [range 2.1–16.5 mmol/L]).
• Many dogs in the survivor group had very high lactate concentration (range 0.5–20 mmol/L).

• Retrospective study.
• Number of dogs in some outcome groups were quite small (power analysis not reported).
• Some cases were excluded based on presumptive diagnosis of GDV.
• High percentage of dogs were euthanised preoperatively (without intent to treat and without radiographic confirmation of diagnosis) which accounted for the majority of mortality in this study, with potential for bias of results.
• Method of lactate measurement not specified.

• Dogs with a physical and radiographic diagnosis of gastric dilatation volvulus (GDV) and in which orogastric decompression and gastric lavage had been performed.
• Mixed population represented.
• Lactate measurement technique not specified.

• Intravenous fluid resuscitation, gastric decompression, gastric lavage.
• Surgical correction of GDV (immediate or as a staged procedure as described below).

• Survival vs non-survival correlated with initial plasma lactate concentration.
• Survival vs non-survival correlated with percentage change in plasma lactate concentration following gastric decompression and stabilisation.
• Survival vs non-survival correlated with performing surgical correction for GDV under the same anaesthesia as for decompression, or as a staged procedure with corrective surgery being performed some hours later following gastric decompression and medical stabilisation prior to exploratory laparotomy.

• Non-survivors had a mean initial plasma lactate concentration of 7.3 mmol/L (range 2.7–11.1 mmol/L) while survivors had a mean initial plasma lactate concentration of 3.2 mmol/L (range 1.2–7.8 mmol/L) (p <0.001).
• Larger percentage decreases in plasma lactate concentration following decompression and stabilisation were associated with higher likelihood of survival to discharge (p <0.03) – full data not provided but 14/15 dogs that survived decreased their plasma lactate by ≥ 50%.

• Retrospective study.
• Dogs were excluded if orogastric intubation was not included as part of gastric decompression.
• Dogs were excluded if, following decompression and stabilisation, they were discharged or transferred to another hospital for corrective surgery.
• Timing of lactate measurement post stabilisation was not standardised.