UBC Faculty Research and Publications

Outcomes of post-cardiac surgery patients with persistent hyperlactatemia in the intensive care unit:… Mak, Nicole T J J; Iqbal, Sameena; de Varennes, Benoit; Khwaja, Kosar Feb 24, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-13019_2016_Article_411.pdf [ 730.06kB ]
JSON: 52383-1.0361958.json
JSON-LD: 52383-1.0361958-ld.json
RDF/XML (Pretty): 52383-1.0361958-rdf.xml
RDF/JSON: 52383-1.0361958-rdf.json
Turtle: 52383-1.0361958-turtle.txt
N-Triples: 52383-1.0361958-rdf-ntriples.txt
Original Record: 52383-1.0361958-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessOutcomes of post-cardiac surgery patientswith persistent hyperlactatemia in theintensive care unit: a matched cohort studyNicole T. J. J. Mak1, Sameena Iqbal2, Benoit de Varennes3 and Kosar Khwaja4*AbstractBackground: Higher morbidity and mortality rates are seen amongst patients presenting with hyperlactatemia inthe postoperative period. The purpose of this study was to determine the relationship between persistent elevations inlactate and poor ICU outcome in post-cardiac surgery patients.Methods: This was a retrospective matched cohort analysis of cardiac surgery patients undergoing bypass and/or valvesurgery in a university hospital centre. Selection criteria were: cardiac bypass and/or valve surgery; admission to the ICUfor > 24 h postoperatively; and peak lactate≥ 3.0 mmol/L. Hyperlactatemic patients were matched to 2 normolactatemicpatients. Multivariable conditional logistic regression was used to determine predictors of hyperlactatemia and mortality.Results: Four hundred sixty-nine post-cardiac surgery patients were admitted to the ICU for > 24 h. 144 of thesepatients had an arterial blood lactate≥ 3.0 mmol/L. Amongst the mortalities, 78.9 % presented with hyperlactatemia.Independent risk factors predictive of a lactate ≥3.0 mmol/L were preoperative IABP insertion (RR 2.8, CI 1.1–7.2) andpostoperative acute kidney injury (RR 3.2, CI 2.1–5.4). Patients whose lactate concentrations continued to increase >30 hpostoperatively were more likely to die (RR 8.44 CI 2.50–28.53).Conclusions: The persistence of hyperlactatemia is a more important determinant of postoperative outcome than theabsolute value of the peak lactate concentration. A simple postoperative lactate washout does not sufficiently explain thislactate accumulation. Mortality is proposed to be secondary to a state of ongoing hypoperfusion.Keywords: Lactate, Surgery, Cardiac, Surgical intensive care, Acidosis, Lactic, Postoperative careBackgroundA product of anaerobic metabolism, lactate is a bio-marker and indicator for tissue hypoperfusion and oxy-gen debt [1, 2]. In anaerobic metabolism, pyruvate isconverted to lactate so that, in conditions of hypoxia,energy may continue to be derived from glucose. Patho-logical causes of lactic acidosis include those conditionscausing tissue hypoxia: pulmonary disease leading tolow PO2; circulatory shock with decreased oxygen deliv-ery; and decreased hemoglobin and oxygen carryingcapacity [3, 4]. Lactic acidosis secondary to hypoxia orhypoperfusion is known as “type A” lactate acidosis. Arise in lactate may also be of the “B” type, which is dueto a lack of lactate clearance by the liver and kidneys.Lactate concentrations are measured regularly in theintensive care unit (ICU) and elevated blood lactate is acommon finding post-cardiac surgery. In the critically illpatient, a normal blood lactate is less than 2 mmol/L.High postoperative serum lactate levels have been previ-ously associated with an increased risk of postoperativemortality and poor outcome in both adult and pediatriccardiac surgery populations [2–10]. Lactate is also abiomarker and potential prognostic tool in the acute caresetting. In adult surgical patients, greater mortality ratesare seen amongst patients in whom lactate levelsnormalize more slowly. Non-normalizing lactate levelshave been associated with a 100 % postoperative mortalityrate [5].* Correspondence: kosar.khwaja@mcgill.ca4Departments of Surgery and Critical Care Medicine, McGill University HealthCentre, 1650 Cedar Ave. L9.411, Montréal, QC H3G 1A4, CanadaFull list of author information is available at the end of the article© 2016 Mak et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 DOI 10.1186/s13019-016-0411-5In this matched cohort study, we describe postopera-tive outcome in a population of hyperlactatemic cardiacsurgery patients. An analysis of the relationship betweenhyperlactatemia, time to peak lactate concentrations andpostoperative mortality was performed. The predictorsof elevated lactate were determined and the value of ser-ial lactate measurements evaluated.MethodsStudy population and matched controlsA retrospective chart review was conducted on patientsundergoing cardiac surgery at the Royal Victoria Hospital(Montreal, Quebec) from January 2011 to December 2011,inclusive. Inclusion criteria were: 1) elective or emergentcoronary artery bypass and/or valve surgery, 2) admissionto the ICU for postoperative care > 24 h in duration, 3)peak arterial blood lactate ≥ 3.0 mmol/L (Fig. 1). Eachhyperlactatemic patient was matched to 2 control patientsselected from cardiac surgery patients whose operationoccurred between January 2010 and December 2012. Casesand controls were matched for sex, age, surgery type,Parsonnet score (a risk estimation score of patient pre-operative factors including diabetes, left ventricular ejec-tion fraction, re-operative status, and dialysis-dependence)and 2 preoperative comorbidities associated with mortalityrisk—renal failure and peripheral vascular disease. Patientswithout matched controls were excluded from analysiswhere appropriate.Data collectionPreoperative data including comorbidities and risk strati-fication scores were obtained from the hospital’s cardiacsurgery database. Postoperative data was collected fromthe intensive care unit database and the hospital’s clin-ical information system database. Additionally, patientspresenting with lactate ≥ 3.0 mmol/L in the ICU under-went a chart review. Postoperative data collected was lim-ited to events occurring during the ICU stay. Mortalitywas defined as death during the cardiac surgery hospitalstay. Serial postoperative lactate values for each patientwith arterial blood lactate values ≥ 3.0 mmol/L weretrended to determine the time to the peak lactate value.This was defined as the number of hours elapsed betweenthe surgery end-time to the time of peak sample collec-tion. Preoperative baseline serum creatinine concentra-tions were recorded to determine the occurrence of newacute kidney injury using the Acute Kidney InjuryNetwork (AKIN) stage 1 criteria. Lastly, diagnostic inter-ventions performed in the ICU such as CT, MRI, echocar-diogram, or ultrasound were noted.Statistical methodsStatistical analyses were performed using the SAS System(SAS 9.3 Institute, Cary, NC). Patient characteristics weresummarized using proportions, medians, and ranges asappropriate. The data was analysed in two parts. First, acohort study of the patients with lactate ≥3.0 mmol/L wasconducted to determine outcomes. This was followed by amatched cohort study to assess risk factors for hyperlacta-temia and mortality.For the cohort study, student’s t-tests or Wilcoxon’srank-sum and Kruskal-Wallis tests were used to compareclinical variables of survivors and non-survivors. Variableswith univariate significance were entered into the Cox Pro-portional Hazard regression model which assessed the ef-fect of time to the development of abnormally high serumlactate levels while controlling for possible confounders. Acategorical variable was created for the 90th percentile fortime to peak lactate concentrations (30 h). Kaplan MeierFig. 1 Selection criteria of cases (a) and controls (b) for the matched cohort studyMak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 2 of 8curves were created to depict the survival curves for thosereaching peak lactate before and after 30 h.In the matched cohort analysis, Wilcoxon’s rank-sumand Kruskal-Wallis tests were used to compare clinicalvariables between normolactatemic and hyperlactatemicpatients as well as between survivors and non-survivors.The individually matched cases and controls were ana-lyzed using conditional logistic regression through theCox Proportional Hazard model with strata. Variables withunivariate significance were entered into the model andpredictors for in-hospital mortality were determined.Generalized estimating equations for binary data wereused to assess the association of elevated serum lactatelevels with postoperative intra-aortic balloon pump(IABP).Data are presented as the following, as appropriate:number (proportion in percentage), median (range), hazardratio (95 % CI). A p-value < 0.05 was considered significant.This study was approved by the McGill UniversityHealth Centre ethics review board. No patient consentwas required.ResultsPatient demographicsPostoperative lactate ≥ 3.0 mmol/L was observed in 195cases out of the 428 patients reviewed. The proceduresperformed in the patient population included 279 coron-ary artery bypass grafts (CABG), 29 aortic valve repairs/re-placements (AVR), 23 mitral valve repairs/replacements(MVR), 45 CABG +AVR, 7 CABG +MVR, and 4 MVR +tricuspid valve repair/replacement. The matched groupshad similar preoperative characteristics for the variablesfor which they were matched. Baseline demographics andpreoperative comorbidities of cases and controls are foundin Table 1. Data was non-normally distributed and there-fore the median was used to describe and compare caseswith controls. Prevalence of preoperative cardiogenicshock and IABP insertion was higher in patients withpostoperative peak lactate ≥3.0 mmol/L. The matchedcontrols had a higher frequency of chronic obstructivepulmonary disease and diabetes mellitus.Postoperative outcomesA summary of postoperative complications is presented inTable 1. The median peak postoperative arterial bloodlactate amongst the cases was 4.6 mmol/L. That of thematched controls was 2.1 mmol/L. Over 15 % of casesreturned to the operating room (OR) for further interven-tion: 8 for hemorrhage/re-exploration, 4 for pericardialeffusion, 2 for perforated bowel, 2 for bowel ischemia/ob-struction, 2 for peripheral ischemia and 1 for left ventricu-lar assist device. By contrast, only 1.8 % of controlsreturned to the OR for hemorrhage/re-exploration. Theuse of IABP was more frequently employed in those withhyperlactatemia (p < 0.0001). A significantly higher occur-rence of complications such as cardiogenic shock (6.3 %,p = 0.002), cardiac hemorrhage (6.3 %, p = 0.01), and acutekidney injury (50.4 %, p < 0.0001) were also observed.Renal dysfunction was further characterized by comparingserum creatinine levels. Preoperative baseline serum cre-atinine levels were similar between both groups but peakpostoperative creatinine was found to be higher in thosepresenting with lactate >3.0 mM. No significant differencewas found with respect to the usage of angiography butTable 1 Comparison of demographics and perioperative courseof cases and controlsCases(n = 144)Controls(n = 280)p-valuePreoperative characteristicsGender (male) 93 (64.6 %) 181 (64.6 %) 1.00Age (years) 70 (43–89) 70 (42–89) 0.81Parsonnet score 18.3 (0–49.0) 16.5 (0–52.5) 0.28Renal failurea 22 (15.3 %) 29 (10.4 %) 0.16Severe peripheral vasculardisease13 (9 %) 17 (6.1 %) 0.32Cardiogenic shockb 5 (3.5 %) 1 (0.4 %) 0.02Preoperative intra-aorticballoon pump16 (11.1 %) 10 (3.6 %) 0.004Chronic obstructivepulmonary diseasec10 (6.9 %) 48 (17.1 %) 0.004Diabetesd 38 (26.4 %) 103 (36.8 %) 0.04Postoperative characteristicsPeak lactate (mmol/L) 4.6 (3.0–18.0) 2.1 (0.8–2.9) <0.0001Peak serum creatininelevel (mmol/L)127 (66–840) 100 (39–1013) <0.0001Intra-aortic balloon pump 29 (20.1 %) 14 (5 %) <0.0001Computed tomographychest /abdomen4 (2.8 %) 0 0.01Return to operating room 22 (15.3 %) 5 (1.8 %) <0.0001Cardiogenic shock 9 (6.3 %) 2 (0.7 %) 0.002Post-operative cardiachemorrhage9 (6.3 %) 4 (1.4 %) 0.01Acute kidney injury (AKIN 1) 68 (50.4 %) 73 (26.8 %) <0.0001Ischemic bowel 1 (0.7 %) 0 0.34Length of ICU admission(days)3 (1–96) 1 (1–29) <0.0001ICU readmission 12 (8.3 %) 13 (4.7 %) 0.13Mechanical ventilation (days) 0.89 (0.14–75.5) 0.50 (0.09–11.9) <0.0001Mortality 15 (10.4 %) 4 (1.4 %) <0.0001Cases presented with peak lactate ≥3.0 mmol/L in the postoperative ICUadmission while controls had a peak lactate <3.0 mmol/L. Data is expressedas: number of patients (frequency in percentage) or median (range)aAcute or chronic renal failurebWith urinary output <10 cc/hrcChronic obstructive pulmonary disease on medical treatmentdDiabetes on insulin or oral hypoglycemic agentsMak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 3 of 8more CT scans of the chest and/or abdomen were con-ducted for the hyperlactatemic patients.Among the cases and controls, length of ICU stay,time on mechanical ventilation and hospital mortalitywere measured (Table 1). With respect to these mea-sures, hyperlactatemic patients had poorer outcomes,having a longer ICU stay (3 vs. 1 days, p < 0.0001),greater time on mechanical ventilation (0.89 vs 0.50 days,p < 0.0001), and higher hospital mortality (p < 0.0001).The mortality rate amongst hyperlactatemic patients was10.4 % compared to 1.4 % in the controls. No differenceswere found in the length of hospital admission or thefrequency of ICU readmission.A comparison between survivors and non-survivors wasconducted in the cohort as well as in the matched-controlanalyses. Comparative analysis revealed that, among thecases of hyperlactatemia, non-survivors had a mean Par-sonnet score of 31.5, compared to a score of 16.75 in survi-vors (Table 2). Non-survivors were more often patientswith pre-existing renal failure, chronic obstructive pulmon-ary disease (COPD), hypertension and heart failure. Emer-gent surgery, longer cardiopulmonary bypass (CPB) timeand requirement for postoperative IABP were also ob-served to be significantly increased in non-survivors. Themean CPB time amongst those who died was 153 mincompared to 110 min in those that survived (p = 0.01).Death was associated with a higher peak lactate (10.2 vs4.4 mmol/L, p = 0.0002) and a delayed time to peak lactate.The 90th percentile for time to peak lactate was 30 h. Asignificantly greater proportion of non-survivors surpassedthe 90th percentile: in 64.7 % of mortalities, lactate peakwas reached at >30 h (p < 0.0001). On average, hyperlacta-temic non-survivors attained their peak lactate in 37.6 hcompared to 7.5 h in survivors (p < 0.0001). Again, differ-ences were noted with respect to renal function. Acute kid-ney injury was more frequent in non-survivors. Peakcreatinine and percentage rise in postoperative creatininewere also greater.Comparative analysis of non-survivors and survivorsin the matched-control population showed that pre-operative renal failure, severe peripheral vascular diseaseand reoperation were associated with mortality while ageand congestive heart failure approached significance(Table 3). The majority of patients who died presentedwith hyperlactatemia (78.9 %) and the 28-day mortalityamongst hyperlactatemic patients was 5.6 % (8/144 pa-tients). Mortality in the study population was associatedwith a higher incidence of postoperative surgical inter-ventions, IABP and complications such as cardiogenicshock, cardiac hemorrhage, and renal dysfunction.Table 2 Comparison of perioperative characteristics of mortality and survival amongst hyperlactatemic patientsNon-survivors (n = 17) Survivors (n = 127) p-valuePre-operative characteristicsGender (male) 9 (52.9 %) 84 (66.1 %) 0.29Age (years) 75 (56–87) 70 (43–89) 0.12Parsonnet score 31.5 (17.0–49.0) 16.8 (0–46.5) <0.0001Parsonnet score greater than 31 20 (15.8 %) 9 (52.9 %) 0.001Operation status (emergency) 3 (17.7 %) 14 (11.5 %) 0.03Surgery type 0.36Dialysis 3 (17.7 %) 3 (2.4 %) 0.02Chronic Obstructive Pulmonary Disease 4 (23.5 %) 6 (4.7 %) 0.02Congestive Heart Failure 11 (64.7 %) 50 (39.4 %) 0.05Post-operative characteristicsCardiopulmonary Bypass time (min) 153 (83–214) 110 (25–302) 0.01Post-operative Intra-aortic Balloon Pump 8 (47.1 %) 21 (16.5 %) 0.003Time to peak lactate >30 h 11 (64.7 %) 3 (2.4 %) <0.0001Time to peak lactate (hours) 37.6 (1.3–385) 7.5 (0.25–755.5) <0.0001Peak arterial lactate (mmol/L) 10.2 (2.4–18) 4.4 (1.7–13.1) 0.0002Total Mechanical Ventilation Days 3.14 (0–76.5) 0.89 (0.14–44.1) 0.0001Acute kidney injury 16 (100 %) 64 (51.6 %) 0.0002Peak creatinine (μmol/L) 309 (107–738) 119 (66–840) <0.0001Postoperative rise in creatinine (%) 30 (−80 to 254) 164 (31 to 660) <0.0001Cases presented with peak lactate ≥3.0 mmol/L in the postoperative ICU admission. Data is expressed as: number of patients (frequency in percentage) ormedian (range)Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 4 of 8Predictors of hyperlactatemia and mortalityConditional logistic regression was used to determine pre-dictors of hyperlactatemia and mortality. Two factors pre-dictive of a peak postoperative lactate ≥3.0 mmol/L werefound: insertion of preoperative IABP and postoperativeacute kidney injury. Adjusted hazard ratios are presentedin Table 4. Patients requiring preoperative IABP insertionwere 2.8 times more likely to develop hyperlactatemia(95 % CI 1.1–7.2). Acute kidney injury was associated witha hazard ratio of 3.2 (95 % CI 2.1–5.4).Despite the association between hyperlactatemia andmortality on univariate analysis, the logistic regressionmodel conducted did not find lactate ≥3.0 mmol/L to bea predictor of death. However, when analysing hyperlac-tatemic patients as a cohort, calculated hazard ratiosshowed that a delayed peak in lactate was strongly asso-ciated with death (Table 5). Patients who exceeded the90th percentile for time to peak lactate were 8.44 (95 %CI 2.50–28.53) times more likely to die. Survival curvesillustrate that overall survival outcomes were better inthose who had a shorter duration of lactate rise (Fig. 2).Other risk factors for mortality were percentage increasein serum creatinine and Parsonnet score.DiscussionThe relationship between hyperlactatemia and poor out-come has been established in adult and pediatric popula-tions. Mortality risk has been associated with bothintraoperative and postoperative hyperlactatemia [2–17].Lactate levels between 2.0 and 4.4 mmol/L have been re-ported to be associated with increased postoperativemorbidity and mortality. Debate exists in the literatureregarding which lactate measurement is most indicativeof poor outcome. Some have measured lactate concen-trations under CPB or peak postoperative lactate whileothers have found that lactate levels upon ICU admis-sion were predictive of outcome [5, 7–13, 15–19]. Otherstudies in medical and surgical critical care have re-ported that better outcomes are observed in the pres-ence of effective lactate clearance [6, 13, 19–23]. In sum,different parameters have been used to characterize therelationship between lactate and poor outcome. Further-more, evidence exists to suggest that survivors may bein a metabolic state more favorable for early resolutionof the hyperlactatemic state. It may be that, in criticallyill patients with multi-organ dysfunction, duration oflactic acidosis has greater predictive value than that ofinitial lactate values at first presentation.Lactate concentration is determined by both the pro-duction and clearance of lactate. An elevated lactate maybe secondary to a previous state of anaerobic metabol-ism that has not yet been cleared. Therefore, the actualmetabolic state of the patient may not be determined bya sole reliance on a lactate concentration. Given thathepatic and renal clearance is often impaired in theTable 3 Comparison of survival and non-survival amongst casesand controls: demographics and perioperative characteristicsNon-survivors(n = 19)Survivors(n = 405)p-valuePreoperative characteristicsGender (male) 11 (57.9 %) 263 (64.9 %) 0.62Age (years) 75 (56–87) 70 (42–89) 0.05Parsonnet score 30.5 (16.5–49) 16.5 (0–52.5) <0.0001Congestive heart failure 12 (63.2 %) 170 (41.9 %) 0.06Renal failurea 9 (47.4 %) 42 (10.4 %) <0.0001Severe peripheral vasculardisease5 (26.3 %) 25 (6.2 %) 0.007Reoperation 3 (15.8 %) 8 (2 %) 0.005Chronic obstructivepulmonary diseaseb7 (36.8 %) 51 (12.6 %) .008Postoperative characteristicsPeak arterial lactate≥3.0 mmol/L15 (78.9 %) 129 (31.8 %) <0.0001Peak creatinine μmol/L 127 (66–840) 100 (39–1013) <0.0001Intra-aortic balloon pump 6 (31.6 %) 37 (9.2 %) 0.008CT chest /abdomen 2 (10.5 %) 2 (0.5 %) 0.01Return to OR 5 (26.3 %) 22 (5.4 %) 0.005Cardiogenic shock 5 (26.3 %) 6 (1.5 %) <0.0001Post-operative cardiachemorrhage2 (10.5 %) 11 (2.7 %) 0.01Acute kidney injury (AKIN 1) 16 (84.2 %) 125 (32.2 %) <0.0001Non-obstructive ileus 1 (5.5 %) 2 (0.5 %) 0.13Ischemic bowel 1 (5.5 %) 0 0.05Data is expressed as: number of patients (frequency in percentage) ormedian (range)aAcute or chronic renal failurebCOPD on medical treatmentTable 4 Risk factors for lactate ≥3.0 mmol/L during thepostoperative ICU admissionVariable Unadjusted OR(95 % CI)Adjusted OR(95 % CI)Diabetes status 0.53 (0.33–0.86) 0.47 (0.28–0.77)Preoperative intra-aorticballoon pump3.0 (1.2–7.3) 2.8 (1.1–7.2)Acute kidney injury (AKIN 1) 3.1 (2.0–4.7) 3.2 (2.1–5.1)n = 424Table 5 Adjusted and unadjusted hazard ratios for mortality inpatients with peak postoperative lactate ≥3.0 mmol/LAdjusted HR(95 % CI)Unadjusted HR(95 % CI)Time to peak lactate≥ 30 h 8.44 (2.50–28.53) 7.13 (2.17–23.45)Percent rise in serum creatinine 1.72 (1.15–2.58) 1.40 (1.03–1.90)Parsonnet score > 31 5.21 (1.07–25.44) 0.80 (0.24–2.65)n = 128Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 5 of 8critically ill, it is reasonable to instead observe serial lac-tate measurements. In this study, lactate was followeduntil the peak concentration was reached. That is, mea-surements were followed so long as there was evidenceof a continued imbalance in the production and clear-ance of lactate. It is expected that time to peak lactate isa better reflection of ongoing active metabolic dysfunc-tion in the critically ill. One other study in the literatureby Ranucci et al. is known to have employed a similarstrategy. Theirs was a study of intraoperative anaerobicmetabolism in patients on cardiopulmonary bypass [17].The results of this study suggest that a single eventof lactate ≥ 3.0 mmol/L is not a reliable predictor ofmortality in post-cardiac surgery patients [24]. Insteada strong relationship was drawn between a late peak-ing lactate and mortality. A peak in lactate at >30 hpostoperatively was a highly significant risk factor fordeath in our population. This risk was independent ofthe absolute concentration of lactate in the blood,CPB time, Parsonnet score and use of postoperativeIABP. Two other risk factors with a significantmortality risk were postoperative rise in creatinineand the Parsonnet score, a risk stratification scorepreviously studied on patients at our institution [25, 26].These findings bring clarity to the approach to interpret-ation of serial lactate measurements in the post-cardiacsurgery patient population.Both cardiogenic and non-cardiogenic causes forhyperlactatemia exist in our population. The multiplelogistic regression models performed here for hyper-lactatemia suggest that kidney dysfunction and re-quirement for insertion of a preoperative IABP areassociated with increased risk of postoperative hyper-lactatemia. The pathophysiology of the hyperlactate-mic state is primarily one of hypoperfusion, resultingin anaerobic metabolism and lactic acid production.Meregalli et al. argued that circulatory dysfunction ispresent even in critically ill surgical patients withoutevidence of shock. It can be inferred that thepresence of hyperlactatemia indicates “occult hypoper-fusion” in certain patients, thus explaining their in-creased mortality [23]. Interestingly, in our study,acute kidney injury in the postoperative period, de-fined by the 2007 Acute Kidney Injury Network stage1 criteria, is highly correlated with hyperlactatemia.From this, we learn that hyperlactatemia is not only aFig. 2 Survival curve for cases of lactate≥ 3.0 mmol/L showing survival probability for patients attaining peak lactate in≥ 30 h (triangles) and thoseattaining their peak lactate in < 30 h (circles)Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 6 of 8result of decreased oxygen delivery, but that it is alsoaffected by end organ dysfunction. Interpretation oflactate trends therefore should take into account suchnon-cardiogenic factors.In the cardiac surgery population specifically, perfusiondeficits are created while on CPB. As core temperature isincreased and CPB is weaned, perfusion of organsbecomes inadequate for demand, resulting in lactic acid-osis [27]. In a prospective study, length of time on CPBhas been previously suggested to increase the likelihood ofdeveloping a postoperative lactatemia ≥ 5.0 mmol/L [28].While the perfusion deficits developed during CPB areundoubtedly contributing to the cardiac surgery patient’sanaerobic state in the postoperative period, our resultsshow that CPB time does not have an independent effectof significance on the development of a postoperativehyperlactatemia ≥ 3.0 mmol/L. This lack of detection of ameasurable effect of CPB time on lactate elevation may bedue to our lower threshold value for hyperlactatemia.The lactate produced intraoperatively while on CPBcontinues to be detected in the serum postoperatively as itis cleared by the kidneys and liver. In addition to this post-operative “washout phenomenon”, perfusion deficits maypersist into the recovery period, leading to lactate persist-ence. We do observe that this development of late hyper-lactatemia is associated with higher mortality risk. Thecontinued lactate production post-CPB is not sufficientlyexplained by a post-CPB washout phenomenon. A shortduration of lactate accumulation and an early peak inserum lactate—in other words, a washout—likely engenderless morbidity and mortality because of the absence of on-going perfusion deficits. We therefore deduce that a hyper-lactatemia of significance is the result of a state of ongoinghypoperfusion, manifesting as a longer duration of lactateaccumulation and a greater rise in serum creatinine.There are limitations to this study. This retrospectivereview included a heterogeneous study population withdifferent cardiac pathologies and procedures. Data onthe use of inotropes, which remain possible confoundersin this study, were not collected. Only patients with anICU stay greater than 24 h were included with the ra-tionale that the hyperlactatemia in these patients was ofclinical significance. A bias in the cohort analysis mayhave resulted from this selection criteria. Note that inthe matched cohort analysis, both the cohort and corre-sponding controls had an ICU admission >24 h.To summarise the findings of this study, we highlight theconcept that the duration of hyperlactatemia is a more im-portant risk factor than the lactate concentration itself. Wepropose that persistent elevations in lactate, for which timeto peak lactate is a surrogate measurement, are due to astate of ongoing hypoperfusion, leading to a significantincrease in mortality risk. This is in contrast to an earlylactate rise which represents a washout phenomenon.ConclusionsCardiac surgery patients presenting with late peakingblood lactate >30 h post-operatively are at increased riskfor poor ICU outcome and mortality. Lactate persistenceis a better measurement of prognosis than peak lactatein post-CPB patients. Patients under a state of persistenthypoperfusion, as opposed to a simple post-CPB lactatewashout, are at risk for an increased frequency of majorpostoperative complications and mortality.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsNM and KK were responsible for study concept and design; data acquisition/analysis; data interpretation; and drafting and revision of the manuscript. SIcontributed substantially to data analysis and interpretation and the writing ofthis manuscript. BD is responsible for the cardiac database and contributed todata interpretation and manuscript review. All authors revised and approvedthe manuscript.AcknowledgmentsWe would like to acknowledge Laura Banici for her assistance with themanagement of ethics approvals and Isabelle Lussier for her assistance withthe ICU database queries. NM received research bursaries from the McGillUniversity Faculty of Medicine (Mach-Gaensslen Foundation, Edward BeattyResearch Award, Clarke McLeod Scholarship). No external funding bodieswere involved in data collection, study design/analysis or in the writing ofthis manuscript.Author details1Division of General Surgery, Vancouver General Hospital, 950 West 10thAvenue, Vancouver, Canada. 2Department of Nephrology, McGill UniversityHealth Centre, Montréal, Canada. 3Division of Cardiothoracic Surgery, McGillUniversity Health Centre, Montréal, Canada. 4Departments of Surgery andCritical Care Medicine, McGill University Health Centre, 1650 Cedar Ave.L9.411, Montréal, QC H3G 1A4, Canada.Received: 19 October 2015 Accepted: 12 January 2016References1. Boldt J, Piper S, Murray P, Lehmann A. Case 2–1999 severe lactic acidosisafter cardiac surgery: Sign of perfusion deficits? J Cardiothorac Vasc Anesth.1999;13(2):220–4.2. Okorie ON, Dellinger P. Lactate: biomarker and potential therapeutic target.Crit Care Clin. 2011;27(2):299–326.3. Mustafa I. Effect of cardiopulmonary bypass on lactate metabolism.Intensive Care Med. 2003;29(8):1279–85.4. Luft FC. Lactic acidosis update for critical care clinicians. J Am Soc Nephrol.2001;12 suppl 1:S15–9.5. McNelis J, Marini CP, Jurkiewicz A, Szomstein S, Simms HH, Ritter G, et al.Prolonged lactate clearance is associated with increased mortality in thesurgical intensive care unit. Am J Surg. 2001;182(5):481–5.6. Attanà P, Lazzeri C, Picariello C, Dini CS, Gensini GF, Valente S. Lactate andlactate clearance in acute cardiac care patients. Eur Heart J AcuteCardiovasc Care. 2012;1(2):115–21.7. Maillet JM. Frequency, risk factors, and outcome of hyperlactatemia aftercardiac surgery. Chest. 2003;123(5):1361–6.8. Kogan A. The impact of hyperlactatemia on postoperative outcome afteradult cardiac surgery. J Anesth. 2012;26(2):174–8.9. Zante B, Kubik M, Reichenspurner H. Lactate level as prognostic factor ofmortality in patients after cardiac surgery. Thorac Cardiovasc Surg. 2010;58:43.10. Mirmohammad-sadeghi M. Relationship between serum lactate levels andmorbidity outcomes in cardiovascular patients after CABG. J Surg Pak Int.2008;13:2.11. Toraman F. Lactic acidosis after cardiac surgery is associated with adverseoutcome. Heart Surg Forum. 2004;7(2):155–9.Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 7 of 812. Yılmaz Ş, Yurtseven N, Tuygun AK, Yavuz Y, Canik S. Hyperlactatemia aftercoronary artery bypass surgery: risk factors and effect on mortality. Turkish JThorac Cardiovasc Surg. 2011;19(1):30–5.13. Hatherill M. Serum lactate as a predictor of mortality after paediatric cardiacsurgery. Arch Dis Child. 1997;77(3):235–8.14. Maarslet L. Lactate levels predict mortality and need for peritoneal dialysisin children undergoing congenital heart surgery. Acta Anaesthesiol Scand.2012;56(4):459–64.15. Demers P. Outcome with high blood lactate levels during cardiopulmonarybypass in adult cardiac operation. Ann Thorac Surg. 2000;70(6):2082–6.16. Ranucci M. Central venous oxygen saturation and blood lactate levelsduring cardiopulmonary bypass are associated with outcome after pediatriccardiac surgery. Crit Care. 2010;14:4.17. Ranucci M. Anaerobic metabolism during cardiopulmonary bypass:predictive value of carbon dioxide derived parameters. Ann Thorac Surg.2006;81(6):2189–95.18. Ranucci M. Hyperlactatemia during cardiopulmonary bypass: determinantsand impact on postoperative outcome. Crit Care. 2006;10:6.19. Marty P. Lactate clearance for death prediction in severe sepsis or septicshock patients during the first 24 hours in Intensive Care Unit: anobservational study. Ann Intensive Care. 2013;3:1.20. Donnino MW. Effective lactate clearance is associated with improvedoutcome in post-cardiac arrest patients. Resuscitation. 2007;75(2):229–34.21. Nguyen H, Bryant HB. Early lactate clearance is associated with improvedoutcome in severe sepsis and septic shock. Crit Care Med. 2004;32(8):1637–42.22. Bakker J. Serial blood lactate levels can predict the development of multipleorgan failure following septic shock. Am J Surg. 1996;171(2):221–6.23. Meregalli A. Occult hypoperfusion is associated with increased mortality inhemodynamically stable, high-risk, surgical patients. Crit Care. 2004;8(2):60–5.24. O’connor E, Fraser J. The interpretation of perioperative lactate abnormalities inpatients undergoing cardiac surgery. Anaesth Intensive Care. 2012;40(4):598–603.25. de Varennes B, Lachapelle K, Cecere R, Ergina P, Shum-Tim D, TchervenkovC, et al. Application of the Parsonnet scoring system for a Canadian cardiacsurgery program. Can J Cardiol. 2007;23(13):1061–5.26. Bernstein AD, Parsonnet V. Bedside estimation of risk as an aid for decision-making in cardiac surgery. Ann Thorac Surg. 2000;69(3):823–8.27. Landow L. Splanchnic lactate production in cardiac surgery patients. CritCare Med. 1993;21(2):84–91.28. Raper RF, Cameron G, Walker D, Bowey CJ. Type B lactic acidosis followingcardiopulmonary bypass. Crit Care Med. 1997;25(1):46–51.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Mak et al. Journal of Cardiothoracic Surgery  (2016) 11:33 Page 8 of 8


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items