Journal of Critical Care 29 (2014) 334–339
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Prognosis of emergency department patients with suspected infection and intermediate lactate levels: A systematic review Michael A. Puskarich, MD, Benjamin M. Illich, Alan E. Jones, MD ⁎ Department of Emergency Medicine, University of Mississippi Medical Center, Jackson, MS
a r t i c l e Keywords: Infection Lactate Prognosis Review
i n f o
a b s t r a c t Purpose: Previous studies have shown a correlation between blood lactate greater than 4.0 mmol/L and mortality in patients with suspected infection in the emergency department (ED), but data are more limited regarding the prognosis of intermediate blood lactate (2.0-3.9 mmol/L), particularly in the absence of hemodynamic instability. We sought to quantify the prognostic significance of intermediate blood lactate levels in ED patients with suspected infection, emphasizing patients without hypotension. Methods: A systematic review of 4 databases was conducted to identify studies using a comprehensive search strategy. All studies performed on adult ED patients with suspected infection and available data on hemodynamics, intermediate lactate levels, and mortality rates were included. Results: We identified 20 potential publications, 8 of which were included. Intermediate lactate elevation was found in 11062 patients with suspected or confirmed infection, 1672 (15.1%) of whom died. Subgroup analysis of normotensive patients demonstrated a mortality of 1561 (14.9%) of 10 442, with rates from individual studies between 3.2% and 16.4%. Conclusion: This systematic review found that among ED patients with suspected infection, intermediate lactate elevation is associated with a moderate to high risk of mortality, even among patients without hypotension. Physicians should consider close monitoring and aggressive treatment for such patients. © 2014 Elsevier Inc. All rights reserved.
1. Purpose In the United States, emergency departments (EDs) treat approximately 570 000 patients with sepsis yearly [1]. In addition, there are 200 000 US deaths attributed to sepsis annually [2], with estimated mortality rates ranging from 15% to 30% based on administrative databases [3]. Despite active research in the field, early administration of bundled care and quantitative resuscitation protocols remains the most consistently beneficial therapy for the treatment of severe sepsis available in the ED. Despite success in the literature, adoption of such protocols has been slow for numerous reasons, including lack of nursing expertise, inability to measure a central venous pressure [4], and inclusion in early goaldirected therapy based on lactate criteria (rather than hypotension) [5]. Furthermore, the placement of central venous lines is accompanied by certain risks including pneumothorax, arterial injury [6], and exposure to the risk of catheter-related blood stream infections [7], a major target of quality improvement initiatives. Furthermore, such protocols increase medical cost per patient, and although such protocols appear to be costeffective in terms of life years gained when considering all patients with septic shock [8,9], it is expected that as the risk of death decreases, costeffectiveness will decrease as well. Thus, it is evident that specific ⁎ Corresponding author. Department of Emergency Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216. E-mail address: [email protected] (A.E. Jones). 0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2013.12.017
identification of at-risk patients for whom bundled care is appropriate should be balanced against the unnecessary risks and expense that accompanies the overtreatment for low-risk individuals. The prognostic value of lactate in ED patients with suspected infection and septic shock is well known [10,11], and increasing levels have been demonstrated to have a curvilinear relationship with mortality [12,13] with no clear stepwise “break” to separate low, moderate, or high risk of death, making proscriptive assessment and treatment guidelines based on these criteria difficult. Based on the original study of early goal-directed therapy [14], the current threshold for initiating early goal directed therapy in the absence of hypotension is a serum lactate greater than 4.0 mmol/L [15]. This number appears to have been selected arbitrarily to define “high” serum lactate, though these patients are clearly at a high risk of death. Observations suggest that patients with a lactate in this range (N 4.0 mmol/L) without hypotension that undergo quantitative resuscitation have a mortality rate not significantly different from those patients requiring vasopressor requirement [16]. What remains unclear, however, is whether patients with an intermediate range lactate, particularly in the absence of hypotension, might similarly benefit from initiation of some type of quantitative resuscitation protocol. As a first step to address this question, we conducted a systematic review to determine the risk associated with a lactate of 2.0 to 3.9 mmol/L (intermediate range), with an emphasis on the subgroup of patients without hypotension, in order to determine the risk of death in this patient population.
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
335
2. Materials and methods
2.4. Outcome measurements
2.1. Search strategy for identification of studies
The primary outcome of interest was 28-day mortality. Other mortality data were also accepted, such as in-hospital mortality and 30-day mortality. To assess the prognostic significance of an intermediate lactate in the setting of normotension on morbidity, we report the rates of progression to either shock or Sequential Organ Failure Assessment (SOFA) score increase of more than 1 if available.
We followed a written protocol that was designed in accordance with recommended guidelines and finalized prior to beginning the study [17]. A preliminary search strategy was developed using exploded Medical Subject Heading terms and keywords involving sepsis, systemic inflammatory response syndrome (SIRS), lactate, and outcome. This strategy was revised and improved upon by a medical librarian. We searched MEDLINE (1950–July 2013), SCOPUS (1996– July 2013), Cochrane Library (2005–July 2013), and CINAHL Plus (1937–July 2013) using a search strategy outlined in Appendix 1. Briefly, our final search strategy included the search terms lactates or acidosis, lactic, and bacterial infections or brain abscess or central nervous system infections or infections, or systemic inflammatory response syndrome and was limited to human studies that reported either prognosis or an outcome assessment (health care). All results were supplemented with a review of their related articles [18].
2.2. Inclusion criteria We considered studies eligible for review, regardless of language or publication type, if they were observational studies of human adults (age N17 years) admitted through the ED, with a diagnosis of SIRS, sepsis, or infection-related diagnosis with available lactate and hemodynamic data. Reviews, correspondence, editorials, and nonhuman studies were excluded; however, their reference lists were screened if relevant to identify further studies for inclusion. We attempted to contact corresponding authors for clarification of missing or incomplete data.
2.5. Subgroup analysis A predefined subgroup of patients with an intermediate lactate and no evidence of hypotension was examined to describe the mortality in this group of patients and compared with those patients with an intermediate lactate and hypotension, when specified. If hemodynamic data were unavailable, we attempted to contact the authors to obtain these data. If the data were unavailable or we failed to receive a response, these patients were excluded from the subgroup analysis. To assess if our choice of outcome measure affected our results, we compare the summary 28- and 30-day mortality rates with the summary in-hospital mortality rate. 2.6. Data analysis Given significant methodological differences between studies, combining the data using meta-analytic techniques was deemed inappropriate. Therefore, we used simple descriptive statistics to present mortality rates as percentages, with accompanying ranges, when appropriate. Summary mortality rates for various groups are reported as a raw sum of total number of deaths divided by total number of patients. 3. Results
2.3. Study selection and data abstraction 3.1. Study selection Two reviewers (B.M.I. and A.E.J.) independently screened the titles and abstracts of identified studies for potential eligibility. Cases of disagreement were resolved by conference between the reviewers. If agreement could not be reached, the full manuscript was obtained for review. The full manuscript of each study that passed the relevance screen was reviewed by one of the investigators (B.M.I.). Study data were abstracted independently by each reviewer using a standardized data collection form. In cases of disagreement in abstracted data, a third reviewer abstracted the data and consensus was reached by conference between the 3 reviewers.
Searching the databases identified 302 potential publications for review (Fig. 1). Reviewing reference lists and related articles found 5 other potential studies. After removing duplicates and screening titles and abstracts, 287 articles were excluded as irrelevant and 20 articles were found to be appropriate for further review. The 20 relevant articles were retrieved in full text. A single reviewer examined all of the references obtained. Twelve of these articles were excluded for failing to meet inclusion criteria. Reasons for exclusion include using a main diagnosis other than sepsis in patient cohorts, lack of
Fig. 1. Search, inclusion, and exclusion flow diagram.
336
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
intermediate lactate data (in some cases, authors were contacted for these data), and the use of a patient database from an already included study. Overall, 8 studies were identified for inclusion [12,18-24]. 3.2. Description of included studies Table 1 summarizes the characteristics of the included studies. Five of the studies were prospective cohorts, and 3 were retrospective. Seven were performed in the United States, and 1 was performed in South Korea. Intermediate lactate elevation was variably defined as 2.0 to 3.9, 2.0 to 4.0, 2.1 to 3.9, and 2.5 to 4.0. Reported mortality rates included 3-, 28-, and 30-day mortality, as well as in-hospital mortality. All studies reporting 3-day mortality also reported 28-day or in-hospital mortality, which was preferentially used for summary descriptive analysis as defined a priori. Six of the 8 studies either reported data to allow for specific subgroup analyses, or authors were able to provide these data upon being contacted. 3.3. Summary of individual included studies Four studies were identified and included in our analysis that evaluated the prognostic value of lactate and included groups of patients with intermediate lactate levels with or without hypotension. Shapiro et al [19] performed a prospective cohort study regarding using serum lactate as a biomarker for predicted sepsis outcome. Inclusion criteria were adults admitted through the ED with an infection-related diagnosis and serum venous lactate obtained. Outcomes reported were 3- and 28-day hospital mortality. A total of 1278 patients were enrolled. Lactate was divided into low (0-2.5 mmol/L), intermediate (2.5-4.0 mmol/L), and high (N4.0 mmol/L) groups. The intermediate group had a mortality rate of 24 (9.0%) of 267 at 28 days. Hemodynamic data were not available for subgroup analysis. Howell et al [21] published a prospective observational cohort study determined to identify the relationship between early serum venous lactate and 28-day mortality in patients with clinically suspected infection. Lactate was similarly divided into low (0-2.5 mmol/L), intermediate (2.5-4.0 mmol/L), and high (N4.0 mmol/L) groups. Author contact was necessary to derive the complete data reported below and in Table 1. Overall mortality was 18 (7.3%) of 246 at 28 days in the intermediate lactate group. Among our predefined subgroup of patients without hypotension, the intermediate lactate group had 12 (5.2%) of 229 deaths at 28 days. Comparatively, those with hypotension (defined as lowest blood pressure in the ED, without details regarding whether this represented transient or refractory hypotension) and intermediate lactate had a mortality of 35.3% (6/17 patients).
Trzeciak et al [20] published post hoc analysis of a prospectively gathered data set in order to determine the usefulness of initial serum lactate as a predictor of hospital mortality. Inclusion criteria were age N 18 years, primary or secondary diagnosis of infection, hospital admission, and a measurement of lactate. In-hospital mortality rate was 25% in the intermediate lactate groups. Intermediate lactate was defined as 2.1 to 3.9 and had a total of 59 (24.8%) of 238 deaths. Specific hemodynamic data were not recorded in this patient registry for further subgroup analysis. Mikkelsen et al [12] published a retrospective cohort study to determine if lactate was an independent predictor of mortality in patients with infection. Patients with severe sepsis were placed into a shock (persistent hypotension despite adequate crystalloid bolus) or nonshock group, based on accepted consensus definitions [25]. Intermediate lactate was defined as 2.0 to 3.9 mmol/L, and the 28-day mortality rates were 16.4% for the intermediate lactate normotension group and 37.3% in patients with intermediate lactate levels and shock. Four further studies were identified that excluded patients with hypotension with the specific goal of documenting the incidence and determining risk factors for the progression to shock among ED patients with suspected infection. Glickman et al [22] published a multicenter prospective cohort study with the goal of identifying risk factors among hemodynamically stable septic patients in order to predict those patients who would progress to shock. Inclusion criteria were patients presenting to the ED with 2 or more SIRS criteria, confirmed infection, and normotension at presentation. The author was contacted directly for information regarding the intermediate lactate group. In patients with a lactate from 2.0 to 3.9 mmol/L, the 30-day mortality was 13 (9.4%) of 138. Song et al [23] published a retrospective cohort study in 2012 specifically addressing patients with sepsis and intermediate lactate level (2.0-4.0 mmol/L). The study was designed to identify risk factors associated with clinical deterioration. Inclusion criteria were patients admitted through the ED with stable hemodynamics and intermediate lactate levels. Of 474 patients enrolled, 48 (10.1%) died in-hospital. Of further interest, 108 (22.7%) of 474 of these progressed to shock with 72 hours of ED arrival. Arnold et al [24] published a multicenter prospective observational study aimed at identifying demographic and physiologic parameters associated with progressive organ failure in patients with “preshock” sepsis, defined as normotension with a lactate of between 2.0 and 3.9 mmol/L. This study used a modified SOFA score as a secondary outcome assessment. The in-hospital mortality rate reported in this study was significantly lower than the others, with only 3 deaths (3.2% mortality). The primary outcome in this study was the composition outcome of a rise in SOFA of more than 1, initiation of vasopressors or mechanical ventilation within 72h of
Table 1 Summary of studies reporting mortality data in patients with an intermediately elevated lactate Author
Study design
Intermediate definitiona
Outcomeb
Overall
Normotensive subgroupc
Shapiro et al [19]
Prospective
2.5-4.0
Prospective Prospective
2.5-4.0 2.1-3.9
Mikkelsen et al [12] Glickman et al [22] Song et al [23]
Retrospective Prospective Retrospective
2.0-3.9 2.0-3.9 2.0-4.0
Arnold et al [24]
Prospective
2.0-3.9
Liu et al [18]
Retrospective
2.0-3.9
24/267 (9.0%) 12/267 (4.5%) 18/246 (7.3%) 59/238 (24.8%) 20/238 (8.4%) 74/415 (17.8%) 13/138 (5.8%) 48/474 (10.1%) 108/474 (22.7%) 3/94 (3.2%) 24/94 (26%) 947/9190 (10.3%) 1433/9190 (15.6%)
No hemodynamic data available
Howell et al [21] Trzeciak et al [20]
28-d 3-d 28-d In-hospital 3-d 28-d 30-d In-hospital Progression to shock In-hospital Progression to SOFA N1 In-hospital 30-d
a b c d
Intermediate lactate as defined by study, mmol/L. Mortality unless otherwise specified. Patients with an intermediate lactate without hypotension: deaths (n)/total (n) (%). Patients with an intermediate lactate with hypotension: deaths (n)/total (n) (%).
Hypotensive subgroupd
12/229 (5.2%) 6/17 (35.3%) No hemodynamic data available 52/317 (16.4%) 13/138 (5.8%) 48/474 (10.1%) 108/474 (22.7%) 3/94 (3.2%) 24/94 (26%) 947/9190 (10.3%) 1433/9190 (15.6%)
22/59 (37.3%) Excluded by study design Excluded by study design Excluded by study design Excluded by study design
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
enrollment, or in-hospital death from any cause. Of the 94 patients enrolled, 24 (26%) met the composite outcome. The largest patient cohort responsible for most patients was recently published by Liu et al [18], who reported a retrospective study that included 9190 patients with a diagnosis of sepsis on admission and an initial lactate between 2.0 and 3.9 mmol/L. The aim of this study was to evaluate the association between initial intermediate lactate levels, lactate clearance, fluid resuscitation, and mortality. Outcomes included in-hospital and 30-day mortality, with 947 (10.3%) and 1433 (15.6%) deaths, respectively. Furthermore, this study subdivided the cohort into those with and without lactate clearance. Most patients (79.5%) showed some improvement in lactate levels by 12 hours. The 30-day mortality rate for this group was 13.3%, whereas it was 24.7% in patients without lactate clearance, consistent with other literature suggesting a favorable prognosis associated with early lactate clearance [26,27]. 3.4. Summary mortality, subgroup mortality, and progression to shock rates Collectively, 11 062 patients were found to meet the inclusion criteria in 8 patient cohorts. Of these 11 062 patients, 1672 (15.1%) had a final outcome of death (in-hospital, 28-day, or 30-day mortality). Further analysis of patients with known normotension (via exclusion of patients with unknown hemodynamics or those patients with known hypotension) left 10 442 patients with intermediate lactate values without hypotension. There were 1561 deaths in this subgroup, for an overall mortality rate of 14.9%, with an individual study range from 3.2% to 16.4%. Although not a primary goal of the study, our review identified a minority of patients with both intermediate lactate levels and hemodynamic instability. Mortality rates in this group of patients were only explicitly described by 2 studies, 35% [21] and 37% [12], respectively. Undoubtedly, there were more hemodynamically unstable patients in 2 other studies [19,20], but lack of hemodynamic data made further analysis impossible. Mortality was similar between those studies reported 28- to 30-day and in-hospital mortality rates (15.3% [1510/9874] vs 10.2% [998/9758]). Hospital length of stay of these patients was only reported in 2 studies, with a median of 4 (intraquartile range, 2-7) [24] and a mean (SD) of 6.1 (7.8) days [18]. Only 1 study reported ICU length of stay, a median of 0 days (intraquartile range, 0-1) [24]. Considering prospective vs retrospective studies, mortality results were similar (11.9% [117/983] vs 10.6% [1069/10 079]), suggesting against significant bias within the retrospective studies and against all of the effects being driven by the single large retrospective cohort. Patients with an intermediate lactate and hypotension had a 28-day mortality that was approximately double (24.3% [28/115 patients]) that of those with normotension. For comparison, studies reporting mortality rates for the high lactate group (N4.0 mmol/L) with normotension observed mortality rates ranging from 26.5% to 31.8%. Finally, the overall rate of progression to shock or SOFA greater than 1 was 23.2% (132/568). 4. Discussion In this systematic review, we used standardized methodology to identify and report the prognostic value of intermediately elevated lactate values in ED patients with suspected infection but without hemodynamic instability and found an overall summary rate of 15%, indicating a nontrivial risk of either in-hospital or 28- to 30-day mortality. This is compared with mortality rates for other acute emergency care conditions such as myocardial infarction (~ 5%) [28], nonmassive pulmonary embolism (15%) [29], and ischemic (12%) stroke [30]. We believe that these data highlight the poor prognosis associated with even a slightly elevated lactate level in the setting of
337
suspected infection and should prompt clinicians to consider more aggressive monitoring and/or therapy for these patients. In the ED, the goal of a physician is to rapidly and accurately assess the severity of a person's illness in order to appropriately choose therapy. Serum lactate has become standard in the assessment of patients with suspected infection [15]. However, despite the data regarding the prognosis associated with intermediate lactate levels (2.0-3.9 mmol/L), a consensus regarding what constitutes appropriately aggressive treatment in this group of patients has yet to emerge. Rivers et al [14] and the Surviving Sepsis Campaign [15] guidelines defined the use of early goal-directed therapy for patients with lactate greater than 4.0 mmol/L even in the absence of persistent hypotension. The data in this review suggest that restricting aggressive resuscitation therapy to only this cohort of patients could potentially miss further at-risk patients in the intermediately elevated lactate group, and we would advocate for more aggressive identification, monitoring, and treatment for these patients. There is no current consensus regarding the therapeutic plan for this group of patients. It would be easy to suggest that merely extending early goal-directed therapy and quantitative resuscitation protocols to these patients is indicated given their similar pathophysiology and relatively high mortality rates. However, such a strategy would require placement of a large number of central venous catheters with all of their attendant risks, in a group of patients that may or may not require them. Although the subgroup of patients with both hypotension and an intermediately elevated lactate would qualify for quantitative resuscitation strategies [15] and may require vasopressors if unresponsive to fluid resuscitation, most patients included in this review had no clinical indication for vasopressors (ie, refractory hypotension), at least not initially in the ED. Placement of central venous catheters in this group of patients would therefore be solely for the purposes of monitoring central venous pressure, the value of which has been questioned [31]; monitoring central venous oxygen saturation, which is of unproven value in this group of patients; or “prophylactically” in the event that the patient might later deteriorate and require vasopressors. Therefore, therapeutic options for these patients at high risk of clinical deterioration remain of uncertain benefit at this time. At a minimum, given data suggesting a high rate of progression to shock [23,24], close monitoring would seem indicated in these patients so that appropriate therapy can be initiated immediately if the patient progresses to septic shock. Based on data suggesting the association between administration of antibiotics and time of shock onset [32-34], it would also seem reasonable to continue to advocate for aggressive, early antibiotic administration as soon as feasible in this group of patients with a high risk of death, although we are aware of no data that clearly demonstrate the time sensitivity of antibiotic administration in patients without hemodynamic shock. Intravenous fluid resuscitation, given its cornerstone place in the resuscitation of septic shock, would seem to be another reasonable therapeutic option, although it remains of unproven benefit in this situation. Finally, in the setting of normotension where resuscitation to specific physiologic end points such as blood pressure is not an option, lactate clearance (decrease in lactate levels over time) may prove to be a reasonable therapeutic target, given its prognostic value [10,35] and use in the setting of septic shock [15]. This issue is of particular relevance and may be an area for future research, as the most recent Surviving Sepsis Campaign guidelines now given a 2C recommendation for the use of lactate normalization (the decrease of an abnormal lactate to a subsequent normal lactate) as a therapeutic target during the first 6 hours of resuscitation. Although 2 randomized clinical trials have demonstrated the use of lactate clearance as a resuscitation parameter, the 2C recommendation results partly due to the fact that these studies used different lactate clearance goals (10% and 20%, respectively) and that no interventional trial to date has evaluated the use of lactate normalization as a therapeutic target.
338
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
Since the publication of these guidelines, one observational study has demonstrated the prognostic but not therapeutic value of lactate normalization [36]. As with the other aforementioned therapies, targeting lactate normalization remains unstudied among patients with intermediate lactate levels and normotension and requires further study. 4.1. Strengths and limitations Using standardized predetermined methodology, we performed a comprehensive literature review, more than 11 000 with an intermediate lactate level and more than 10 000 with stable hemodynamics. These data give us insight into the prognostic value of intermediate lactate values in the setting of suspected infection. Although our search was broad and encompassed numerous medical libraries, it is possible that conference abstracts and other unpublished data from the “gray” literature were not identified, limiting our results. Furthermore, given the significant methodologic differences between studies, it would not have been appropriate to perform a metaanalysis in this setting, which limits the interpretation of any pooled data. Finally, a systematic review is limited by the quality of the published literature, and any selection bias or differential treatment for these patients for whom no clear standard of care exists may have affected our results and reported mortality rates. 5. Conclusion In this systematic review, we found that among ED patients with suspected infection, intermediate lactate elevation is associated with a moderate to high risk of mortality compared with many other acute care diseases, even among patients without hypotension. Physicians should be aware of the poor prognosis of this group of patients, should monitor them closely, and give consideration to more aggressive treatment in order to prevent further progression to shock or death. Acknowledgments Dr Jones has received salary support from the National Institute of General Medical Sciences (R01 GM103799). Dr Puskarich received salary support from a Career Development Grant from the Emergency Medicine Foundation. Appendix 1. Representative search strategy for the Medline database b1950 to JULY 2013N 1. Lactates/(55420) 2. Acidosis, Lactic/(6720) 3. (((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections (1875648) 4. Systemic Inflammatory Response Syndrome (92052) 5. (Lactates) OR Acidosis, Lactic/(59597) 6. (((((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections)) OR Systemic Inflammatory Response Syndrome (1881924) 7. Prognosis/(11211486) 8. Outcome Assessment (Health Care)/(130252) 9. (Prognosis) OR Outcome Assessment (Health Care) (1186218) 10. (((((Lactates) OR Acidosis, Lactic/)) AND ((((((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections)) OR Systemic Inflammatory Response Syndrome))) AND (((Prognosis) OR Outcome Assessment (Health Care)) (310) 11. Limit (#10) to HUMANS, AGE 19 + (193)
References [1] Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis in United States emergency departments. Crit Care Med 2007;35(8): 1928–36. [2] Angus D, Linde-Zwirble W, Lidicker J, Clermont J, Carcillo G, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29(7):1303–10. [3] Gaieski D, Edwards J, Kallan M, Carr B. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41(5): 1167–74. [4] Carlbom DJ, Rubenfeld GD. Barriers to implementing protocol-based sepsis resuscitation in the emergency department—results of a national survey. Crit Care Med 2007;35(11):2525–35. [5] Mikkelsen M, Gaieski D, Goyal M, Miltiades A, Munson J, Pines JM, et al. Factors associated with nonadherence to early goal-directed therapy in the ED. Chest 2010;138:551–8. [6] Paoletti F, Ripani U, Antonelli M, Nicoletta G. Central venous catheters: observations on the implantation technique and its complications. Minerva Anestesiol 2005;71(9):555–60. [7] Traore O, Liotier J, Souweine B. Prospective study of arterial and central venous catheter colonization and of arterial- and central venous catheterrelated bacteremia in intensive care units. Crit Care Med 2005;33(6): 1276–80. [8] Talmor D, Greenberg D, Howell MD, Lisbon A, Novack V, Shapiro N. The costs and cost-effectiveness of an integrated sepsis treatment protocol. Crit Care Med 2008;36(4):1168–74. [9] Jones AE, Troyer JL, Kline JA. Cost effectiveness of an emergency department based early sepsis resuscitation protocol. Crit Care Med 2011;39:1306–12. [10] Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent J-L. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996;171(2):221–6. [11] Jansen TC, van Bommel J, Bakker J. Blood lactate monitoring in critically ill patients: a systematic health technology assessment. Crit Care Med 2009;37(10): 2827–39. [12] Mikkelsen M, Miltiades A, Gaieski D, Goyal M, Fuchs B, Shah C, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med 2009;37:1670–7. [13] Puskarich M, Kline JA, Summers RL, Jones AE. Prognostic value of incremental lactate elevations in emergency department patients with suspected infection. Acad Emerg Med 2012;19(8):983–5. [14] Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goaldirected therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345(19):1368–677. [15] Dellinger R, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving Sepsis Campaign: international guidelines for the management of severe sepsis and septic shock 2012. Crit Care Med 2013;41(2):580–637. [16] Puskarich M, Trzeciak S, Shapiro N, Heffner AC, Kline JA, Jones AE, et al. Outcomes of patients undergoing early sepsis resuscitation for cryptic shock compared to overt shock. Resuscitation 2011;82(10):1289–93. [17] Cochrane handbook for systematic reviews of interventions 4.2.6 [updated September 2006]. In: Higgins JPT, Green S, editors. Cochrane Library. Issue 4 ed. Chichester, UK: John Wiley & Sons, Ltd.; 2006. [18] Liu V, Morehouse JJ, Soule J, Whippy A, Escobar GJ. Fluid volume, lactate values, and mortality in sepsis patients with intermediate lactate values. Ann Am Thorac Soc 2013;10(5):466–73. [19] Shapiro NI, Howell MD, Talmor D, Nathanson LA, Lisbon A, Wolfe RE, et al. Serum lactate as a predictor of mortality in emergency department patients with infection. Ann Emerg Med 2005;45(5):524–8. [20] Trzeciak S, Dellinger R, Chansky ME, Arnold R, Schorr C, Milcarek B, et al. Serum lactate as a predictor of mortality in patients with infection. Intensive Care Med 2007;33(6):970–7. [21] Howell MD, Donnino M, Clardy P, Talmor D, Shapiro NI. Occult hypoperfusion and mortality in patients with suspected infection. Intensive Care Med 2007;33(11): 1892–9. [22] Glickman SW, Cairns CB, Otero RM, Woods CW, Tsalik EL, Langley RJ, et al. Disease progression in hemodynamically stable patients presenting to the emergency department with sepsis. Acad Emerg Med 2010;17(4):383–90. [23] Song YH, Shin TG, Kang MJ, Sim MS, Jo IJ, Song KJ, et al. Predicting factors associated with clinical deterioration of sepsis patients with intermediate levels of serum lactate. Shock 2012;38(3):249–54. [24] Arnold RC, Sherwin R, Shapiro NI, O'Conner JL, Glaspey L, Singh S, et al. Multicenter obsevational study of the development of progressive organ dysfunction and therapeutic interventions in normotensive sepsis patients in the emergency department. Acad Emerg Med 2013;20(5):433–40. [25] Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31(4):1250–6. [26] Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, et al. Early lactate clearance is associated with improved outcome in severe sepsis and septic shock. Crit Care Med 2004;32(8):1637–42. [27] Puskarich M, Trzeciak S, Shapiro N, Arnold R, Heffner AC, Kline JA, et al. Prognostic value and agreement of achieving lactate clearance or central venous oxygen saturation goals during early sepsis resuscitation. Acad Emerg Med 2012;19(3): 252–8.
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339 [28] Roe MT, Messenger JC, Weintraub WS, Cannon CP, Fonarow GC, Dai D, et al. Treatments, trends, and outcomes of acute myocardial infarction and percutaneous coronary intervention. J Am Coll Cardiol 2010;56(4):254–63. [29] Kucher N, Rossi E, De Rosa M, Goldhaber SZ. Massive pulmonary embolism. Circulation 2006;113:577–82. [30] Lichtman JH, Leifheit-Limson EC, Jones SB, Wang Y, Goldstein LB. 30-Day riskstandardized mortality and readmission rates after ischemic stroke in critical access hospitals. Stroke 2012;43:2741–7. [31] Marik PE, Cavallazzi R. Does the central venous pressure predict fluid responsivenses? An updated meta-analysis and a plea for some common sense. Crit Care Med 2013;41(7):1774–81. [32] Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34(6):1589–96.
339
[33] Puskarich M, Trzeciak S, Shapiro NI, Arnold RC, Horton JM, Studnek JR, et al. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011;39(9):2066–71. [34] Gaieski D, Mikkelsen M, Bank RA, Pines JM, Massone R, Furia FF, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergecy department. Crit Care Med 2010;38(4):1045–53. [35] Arnold RC, Shapiro NI, Jones AE, Schorr C, Pope J, Casner E, et al. Multi-center study of early lactate clearance as a determinant of survival in patients with presumed sepsis. Shock 2009;32(1):36–40. [36] Puskarich M, Trzeciak S, Shapiro NI, Albers AB, Heffner AC, Kliner JA, et al. Whole Blood Lactate Kinetics in Patients Undergoing Quantitative Resuscitation for Severe Sepsis and Septic Shock. Chest 2013;143(16):1548–53.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Prognosis of emergency department patients with suspected infection and intermediate lactate levels: A systematic review Michael A. Puskarich, MD, Benjamin M. Illich, Alan E. Jones, MD ⁎ Department of Emergency Medicine, University of Mississippi Medical Center, Jackson, MS
a r t i c l e Keywords: Infection Lactate Prognosis Review
i n f o
a b s t r a c t Purpose: Previous studies have shown a correlation between blood lactate greater than 4.0 mmol/L and mortality in patients with suspected infection in the emergency department (ED), but data are more limited regarding the prognosis of intermediate blood lactate (2.0-3.9 mmol/L), particularly in the absence of hemodynamic instability. We sought to quantify the prognostic significance of intermediate blood lactate levels in ED patients with suspected infection, emphasizing patients without hypotension. Methods: A systematic review of 4 databases was conducted to identify studies using a comprehensive search strategy. All studies performed on adult ED patients with suspected infection and available data on hemodynamics, intermediate lactate levels, and mortality rates were included. Results: We identified 20 potential publications, 8 of which were included. Intermediate lactate elevation was found in 11062 patients with suspected or confirmed infection, 1672 (15.1%) of whom died. Subgroup analysis of normotensive patients demonstrated a mortality of 1561 (14.9%) of 10 442, with rates from individual studies between 3.2% and 16.4%. Conclusion: This systematic review found that among ED patients with suspected infection, intermediate lactate elevation is associated with a moderate to high risk of mortality, even among patients without hypotension. Physicians should consider close monitoring and aggressive treatment for such patients. © 2014 Elsevier Inc. All rights reserved.
1. Purpose In the United States, emergency departments (EDs) treat approximately 570 000 patients with sepsis yearly [1]. In addition, there are 200 000 US deaths attributed to sepsis annually [2], with estimated mortality rates ranging from 15% to 30% based on administrative databases [3]. Despite active research in the field, early administration of bundled care and quantitative resuscitation protocols remains the most consistently beneficial therapy for the treatment of severe sepsis available in the ED. Despite success in the literature, adoption of such protocols has been slow for numerous reasons, including lack of nursing expertise, inability to measure a central venous pressure [4], and inclusion in early goaldirected therapy based on lactate criteria (rather than hypotension) [5]. Furthermore, the placement of central venous lines is accompanied by certain risks including pneumothorax, arterial injury [6], and exposure to the risk of catheter-related blood stream infections [7], a major target of quality improvement initiatives. Furthermore, such protocols increase medical cost per patient, and although such protocols appear to be costeffective in terms of life years gained when considering all patients with septic shock [8,9], it is expected that as the risk of death decreases, costeffectiveness will decrease as well. Thus, it is evident that specific ⁎ Corresponding author. Department of Emergency Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216. E-mail address: [email protected] (A.E. Jones). 0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2013.12.017
identification of at-risk patients for whom bundled care is appropriate should be balanced against the unnecessary risks and expense that accompanies the overtreatment for low-risk individuals. The prognostic value of lactate in ED patients with suspected infection and septic shock is well known [10,11], and increasing levels have been demonstrated to have a curvilinear relationship with mortality [12,13] with no clear stepwise “break” to separate low, moderate, or high risk of death, making proscriptive assessment and treatment guidelines based on these criteria difficult. Based on the original study of early goal-directed therapy [14], the current threshold for initiating early goal directed therapy in the absence of hypotension is a serum lactate greater than 4.0 mmol/L [15]. This number appears to have been selected arbitrarily to define “high” serum lactate, though these patients are clearly at a high risk of death. Observations suggest that patients with a lactate in this range (N 4.0 mmol/L) without hypotension that undergo quantitative resuscitation have a mortality rate not significantly different from those patients requiring vasopressor requirement [16]. What remains unclear, however, is whether patients with an intermediate range lactate, particularly in the absence of hypotension, might similarly benefit from initiation of some type of quantitative resuscitation protocol. As a first step to address this question, we conducted a systematic review to determine the risk associated with a lactate of 2.0 to 3.9 mmol/L (intermediate range), with an emphasis on the subgroup of patients without hypotension, in order to determine the risk of death in this patient population.
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
335
2. Materials and methods
2.4. Outcome measurements
2.1. Search strategy for identification of studies
The primary outcome of interest was 28-day mortality. Other mortality data were also accepted, such as in-hospital mortality and 30-day mortality. To assess the prognostic significance of an intermediate lactate in the setting of normotension on morbidity, we report the rates of progression to either shock or Sequential Organ Failure Assessment (SOFA) score increase of more than 1 if available.
We followed a written protocol that was designed in accordance with recommended guidelines and finalized prior to beginning the study [17]. A preliminary search strategy was developed using exploded Medical Subject Heading terms and keywords involving sepsis, systemic inflammatory response syndrome (SIRS), lactate, and outcome. This strategy was revised and improved upon by a medical librarian. We searched MEDLINE (1950–July 2013), SCOPUS (1996– July 2013), Cochrane Library (2005–July 2013), and CINAHL Plus (1937–July 2013) using a search strategy outlined in Appendix 1. Briefly, our final search strategy included the search terms lactates or acidosis, lactic, and bacterial infections or brain abscess or central nervous system infections or infections, or systemic inflammatory response syndrome and was limited to human studies that reported either prognosis or an outcome assessment (health care). All results were supplemented with a review of their related articles [18].
2.2. Inclusion criteria We considered studies eligible for review, regardless of language or publication type, if they were observational studies of human adults (age N17 years) admitted through the ED, with a diagnosis of SIRS, sepsis, or infection-related diagnosis with available lactate and hemodynamic data. Reviews, correspondence, editorials, and nonhuman studies were excluded; however, their reference lists were screened if relevant to identify further studies for inclusion. We attempted to contact corresponding authors for clarification of missing or incomplete data.
2.5. Subgroup analysis A predefined subgroup of patients with an intermediate lactate and no evidence of hypotension was examined to describe the mortality in this group of patients and compared with those patients with an intermediate lactate and hypotension, when specified. If hemodynamic data were unavailable, we attempted to contact the authors to obtain these data. If the data were unavailable or we failed to receive a response, these patients were excluded from the subgroup analysis. To assess if our choice of outcome measure affected our results, we compare the summary 28- and 30-day mortality rates with the summary in-hospital mortality rate. 2.6. Data analysis Given significant methodological differences between studies, combining the data using meta-analytic techniques was deemed inappropriate. Therefore, we used simple descriptive statistics to present mortality rates as percentages, with accompanying ranges, when appropriate. Summary mortality rates for various groups are reported as a raw sum of total number of deaths divided by total number of patients. 3. Results
2.3. Study selection and data abstraction 3.1. Study selection Two reviewers (B.M.I. and A.E.J.) independently screened the titles and abstracts of identified studies for potential eligibility. Cases of disagreement were resolved by conference between the reviewers. If agreement could not be reached, the full manuscript was obtained for review. The full manuscript of each study that passed the relevance screen was reviewed by one of the investigators (B.M.I.). Study data were abstracted independently by each reviewer using a standardized data collection form. In cases of disagreement in abstracted data, a third reviewer abstracted the data and consensus was reached by conference between the 3 reviewers.
Searching the databases identified 302 potential publications for review (Fig. 1). Reviewing reference lists and related articles found 5 other potential studies. After removing duplicates and screening titles and abstracts, 287 articles were excluded as irrelevant and 20 articles were found to be appropriate for further review. The 20 relevant articles were retrieved in full text. A single reviewer examined all of the references obtained. Twelve of these articles were excluded for failing to meet inclusion criteria. Reasons for exclusion include using a main diagnosis other than sepsis in patient cohorts, lack of
Fig. 1. Search, inclusion, and exclusion flow diagram.
336
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
intermediate lactate data (in some cases, authors were contacted for these data), and the use of a patient database from an already included study. Overall, 8 studies were identified for inclusion [12,18-24]. 3.2. Description of included studies Table 1 summarizes the characteristics of the included studies. Five of the studies were prospective cohorts, and 3 were retrospective. Seven were performed in the United States, and 1 was performed in South Korea. Intermediate lactate elevation was variably defined as 2.0 to 3.9, 2.0 to 4.0, 2.1 to 3.9, and 2.5 to 4.0. Reported mortality rates included 3-, 28-, and 30-day mortality, as well as in-hospital mortality. All studies reporting 3-day mortality also reported 28-day or in-hospital mortality, which was preferentially used for summary descriptive analysis as defined a priori. Six of the 8 studies either reported data to allow for specific subgroup analyses, or authors were able to provide these data upon being contacted. 3.3. Summary of individual included studies Four studies were identified and included in our analysis that evaluated the prognostic value of lactate and included groups of patients with intermediate lactate levels with or without hypotension. Shapiro et al [19] performed a prospective cohort study regarding using serum lactate as a biomarker for predicted sepsis outcome. Inclusion criteria were adults admitted through the ED with an infection-related diagnosis and serum venous lactate obtained. Outcomes reported were 3- and 28-day hospital mortality. A total of 1278 patients were enrolled. Lactate was divided into low (0-2.5 mmol/L), intermediate (2.5-4.0 mmol/L), and high (N4.0 mmol/L) groups. The intermediate group had a mortality rate of 24 (9.0%) of 267 at 28 days. Hemodynamic data were not available for subgroup analysis. Howell et al [21] published a prospective observational cohort study determined to identify the relationship between early serum venous lactate and 28-day mortality in patients with clinically suspected infection. Lactate was similarly divided into low (0-2.5 mmol/L), intermediate (2.5-4.0 mmol/L), and high (N4.0 mmol/L) groups. Author contact was necessary to derive the complete data reported below and in Table 1. Overall mortality was 18 (7.3%) of 246 at 28 days in the intermediate lactate group. Among our predefined subgroup of patients without hypotension, the intermediate lactate group had 12 (5.2%) of 229 deaths at 28 days. Comparatively, those with hypotension (defined as lowest blood pressure in the ED, without details regarding whether this represented transient or refractory hypotension) and intermediate lactate had a mortality of 35.3% (6/17 patients).
Trzeciak et al [20] published post hoc analysis of a prospectively gathered data set in order to determine the usefulness of initial serum lactate as a predictor of hospital mortality. Inclusion criteria were age N 18 years, primary or secondary diagnosis of infection, hospital admission, and a measurement of lactate. In-hospital mortality rate was 25% in the intermediate lactate groups. Intermediate lactate was defined as 2.1 to 3.9 and had a total of 59 (24.8%) of 238 deaths. Specific hemodynamic data were not recorded in this patient registry for further subgroup analysis. Mikkelsen et al [12] published a retrospective cohort study to determine if lactate was an independent predictor of mortality in patients with infection. Patients with severe sepsis were placed into a shock (persistent hypotension despite adequate crystalloid bolus) or nonshock group, based on accepted consensus definitions [25]. Intermediate lactate was defined as 2.0 to 3.9 mmol/L, and the 28-day mortality rates were 16.4% for the intermediate lactate normotension group and 37.3% in patients with intermediate lactate levels and shock. Four further studies were identified that excluded patients with hypotension with the specific goal of documenting the incidence and determining risk factors for the progression to shock among ED patients with suspected infection. Glickman et al [22] published a multicenter prospective cohort study with the goal of identifying risk factors among hemodynamically stable septic patients in order to predict those patients who would progress to shock. Inclusion criteria were patients presenting to the ED with 2 or more SIRS criteria, confirmed infection, and normotension at presentation. The author was contacted directly for information regarding the intermediate lactate group. In patients with a lactate from 2.0 to 3.9 mmol/L, the 30-day mortality was 13 (9.4%) of 138. Song et al [23] published a retrospective cohort study in 2012 specifically addressing patients with sepsis and intermediate lactate level (2.0-4.0 mmol/L). The study was designed to identify risk factors associated with clinical deterioration. Inclusion criteria were patients admitted through the ED with stable hemodynamics and intermediate lactate levels. Of 474 patients enrolled, 48 (10.1%) died in-hospital. Of further interest, 108 (22.7%) of 474 of these progressed to shock with 72 hours of ED arrival. Arnold et al [24] published a multicenter prospective observational study aimed at identifying demographic and physiologic parameters associated with progressive organ failure in patients with “preshock” sepsis, defined as normotension with a lactate of between 2.0 and 3.9 mmol/L. This study used a modified SOFA score as a secondary outcome assessment. The in-hospital mortality rate reported in this study was significantly lower than the others, with only 3 deaths (3.2% mortality). The primary outcome in this study was the composition outcome of a rise in SOFA of more than 1, initiation of vasopressors or mechanical ventilation within 72h of
Table 1 Summary of studies reporting mortality data in patients with an intermediately elevated lactate Author
Study design
Intermediate definitiona
Outcomeb
Overall
Normotensive subgroupc
Shapiro et al [19]
Prospective
2.5-4.0
Prospective Prospective
2.5-4.0 2.1-3.9
Mikkelsen et al [12] Glickman et al [22] Song et al [23]
Retrospective Prospective Retrospective
2.0-3.9 2.0-3.9 2.0-4.0
Arnold et al [24]
Prospective
2.0-3.9
Liu et al [18]
Retrospective
2.0-3.9
24/267 (9.0%) 12/267 (4.5%) 18/246 (7.3%) 59/238 (24.8%) 20/238 (8.4%) 74/415 (17.8%) 13/138 (5.8%) 48/474 (10.1%) 108/474 (22.7%) 3/94 (3.2%) 24/94 (26%) 947/9190 (10.3%) 1433/9190 (15.6%)
No hemodynamic data available
Howell et al [21] Trzeciak et al [20]
28-d 3-d 28-d In-hospital 3-d 28-d 30-d In-hospital Progression to shock In-hospital Progression to SOFA N1 In-hospital 30-d
a b c d
Intermediate lactate as defined by study, mmol/L. Mortality unless otherwise specified. Patients with an intermediate lactate without hypotension: deaths (n)/total (n) (%). Patients with an intermediate lactate with hypotension: deaths (n)/total (n) (%).
Hypotensive subgroupd
12/229 (5.2%) 6/17 (35.3%) No hemodynamic data available 52/317 (16.4%) 13/138 (5.8%) 48/474 (10.1%) 108/474 (22.7%) 3/94 (3.2%) 24/94 (26%) 947/9190 (10.3%) 1433/9190 (15.6%)
22/59 (37.3%) Excluded by study design Excluded by study design Excluded by study design Excluded by study design
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
enrollment, or in-hospital death from any cause. Of the 94 patients enrolled, 24 (26%) met the composite outcome. The largest patient cohort responsible for most patients was recently published by Liu et al [18], who reported a retrospective study that included 9190 patients with a diagnosis of sepsis on admission and an initial lactate between 2.0 and 3.9 mmol/L. The aim of this study was to evaluate the association between initial intermediate lactate levels, lactate clearance, fluid resuscitation, and mortality. Outcomes included in-hospital and 30-day mortality, with 947 (10.3%) and 1433 (15.6%) deaths, respectively. Furthermore, this study subdivided the cohort into those with and without lactate clearance. Most patients (79.5%) showed some improvement in lactate levels by 12 hours. The 30-day mortality rate for this group was 13.3%, whereas it was 24.7% in patients without lactate clearance, consistent with other literature suggesting a favorable prognosis associated with early lactate clearance [26,27]. 3.4. Summary mortality, subgroup mortality, and progression to shock rates Collectively, 11 062 patients were found to meet the inclusion criteria in 8 patient cohorts. Of these 11 062 patients, 1672 (15.1%) had a final outcome of death (in-hospital, 28-day, or 30-day mortality). Further analysis of patients with known normotension (via exclusion of patients with unknown hemodynamics or those patients with known hypotension) left 10 442 patients with intermediate lactate values without hypotension. There were 1561 deaths in this subgroup, for an overall mortality rate of 14.9%, with an individual study range from 3.2% to 16.4%. Although not a primary goal of the study, our review identified a minority of patients with both intermediate lactate levels and hemodynamic instability. Mortality rates in this group of patients were only explicitly described by 2 studies, 35% [21] and 37% [12], respectively. Undoubtedly, there were more hemodynamically unstable patients in 2 other studies [19,20], but lack of hemodynamic data made further analysis impossible. Mortality was similar between those studies reported 28- to 30-day and in-hospital mortality rates (15.3% [1510/9874] vs 10.2% [998/9758]). Hospital length of stay of these patients was only reported in 2 studies, with a median of 4 (intraquartile range, 2-7) [24] and a mean (SD) of 6.1 (7.8) days [18]. Only 1 study reported ICU length of stay, a median of 0 days (intraquartile range, 0-1) [24]. Considering prospective vs retrospective studies, mortality results were similar (11.9% [117/983] vs 10.6% [1069/10 079]), suggesting against significant bias within the retrospective studies and against all of the effects being driven by the single large retrospective cohort. Patients with an intermediate lactate and hypotension had a 28-day mortality that was approximately double (24.3% [28/115 patients]) that of those with normotension. For comparison, studies reporting mortality rates for the high lactate group (N4.0 mmol/L) with normotension observed mortality rates ranging from 26.5% to 31.8%. Finally, the overall rate of progression to shock or SOFA greater than 1 was 23.2% (132/568). 4. Discussion In this systematic review, we used standardized methodology to identify and report the prognostic value of intermediately elevated lactate values in ED patients with suspected infection but without hemodynamic instability and found an overall summary rate of 15%, indicating a nontrivial risk of either in-hospital or 28- to 30-day mortality. This is compared with mortality rates for other acute emergency care conditions such as myocardial infarction (~ 5%) [28], nonmassive pulmonary embolism (15%) [29], and ischemic (12%) stroke [30]. We believe that these data highlight the poor prognosis associated with even a slightly elevated lactate level in the setting of
337
suspected infection and should prompt clinicians to consider more aggressive monitoring and/or therapy for these patients. In the ED, the goal of a physician is to rapidly and accurately assess the severity of a person's illness in order to appropriately choose therapy. Serum lactate has become standard in the assessment of patients with suspected infection [15]. However, despite the data regarding the prognosis associated with intermediate lactate levels (2.0-3.9 mmol/L), a consensus regarding what constitutes appropriately aggressive treatment in this group of patients has yet to emerge. Rivers et al [14] and the Surviving Sepsis Campaign [15] guidelines defined the use of early goal-directed therapy for patients with lactate greater than 4.0 mmol/L even in the absence of persistent hypotension. The data in this review suggest that restricting aggressive resuscitation therapy to only this cohort of patients could potentially miss further at-risk patients in the intermediately elevated lactate group, and we would advocate for more aggressive identification, monitoring, and treatment for these patients. There is no current consensus regarding the therapeutic plan for this group of patients. It would be easy to suggest that merely extending early goal-directed therapy and quantitative resuscitation protocols to these patients is indicated given their similar pathophysiology and relatively high mortality rates. However, such a strategy would require placement of a large number of central venous catheters with all of their attendant risks, in a group of patients that may or may not require them. Although the subgroup of patients with both hypotension and an intermediately elevated lactate would qualify for quantitative resuscitation strategies [15] and may require vasopressors if unresponsive to fluid resuscitation, most patients included in this review had no clinical indication for vasopressors (ie, refractory hypotension), at least not initially in the ED. Placement of central venous catheters in this group of patients would therefore be solely for the purposes of monitoring central venous pressure, the value of which has been questioned [31]; monitoring central venous oxygen saturation, which is of unproven value in this group of patients; or “prophylactically” in the event that the patient might later deteriorate and require vasopressors. Therefore, therapeutic options for these patients at high risk of clinical deterioration remain of uncertain benefit at this time. At a minimum, given data suggesting a high rate of progression to shock [23,24], close monitoring would seem indicated in these patients so that appropriate therapy can be initiated immediately if the patient progresses to septic shock. Based on data suggesting the association between administration of antibiotics and time of shock onset [32-34], it would also seem reasonable to continue to advocate for aggressive, early antibiotic administration as soon as feasible in this group of patients with a high risk of death, although we are aware of no data that clearly demonstrate the time sensitivity of antibiotic administration in patients without hemodynamic shock. Intravenous fluid resuscitation, given its cornerstone place in the resuscitation of septic shock, would seem to be another reasonable therapeutic option, although it remains of unproven benefit in this situation. Finally, in the setting of normotension where resuscitation to specific physiologic end points such as blood pressure is not an option, lactate clearance (decrease in lactate levels over time) may prove to be a reasonable therapeutic target, given its prognostic value [10,35] and use in the setting of septic shock [15]. This issue is of particular relevance and may be an area for future research, as the most recent Surviving Sepsis Campaign guidelines now given a 2C recommendation for the use of lactate normalization (the decrease of an abnormal lactate to a subsequent normal lactate) as a therapeutic target during the first 6 hours of resuscitation. Although 2 randomized clinical trials have demonstrated the use of lactate clearance as a resuscitation parameter, the 2C recommendation results partly due to the fact that these studies used different lactate clearance goals (10% and 20%, respectively) and that no interventional trial to date has evaluated the use of lactate normalization as a therapeutic target.
338
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339
Since the publication of these guidelines, one observational study has demonstrated the prognostic but not therapeutic value of lactate normalization [36]. As with the other aforementioned therapies, targeting lactate normalization remains unstudied among patients with intermediate lactate levels and normotension and requires further study. 4.1. Strengths and limitations Using standardized predetermined methodology, we performed a comprehensive literature review, more than 11 000 with an intermediate lactate level and more than 10 000 with stable hemodynamics. These data give us insight into the prognostic value of intermediate lactate values in the setting of suspected infection. Although our search was broad and encompassed numerous medical libraries, it is possible that conference abstracts and other unpublished data from the “gray” literature were not identified, limiting our results. Furthermore, given the significant methodologic differences between studies, it would not have been appropriate to perform a metaanalysis in this setting, which limits the interpretation of any pooled data. Finally, a systematic review is limited by the quality of the published literature, and any selection bias or differential treatment for these patients for whom no clear standard of care exists may have affected our results and reported mortality rates. 5. Conclusion In this systematic review, we found that among ED patients with suspected infection, intermediate lactate elevation is associated with a moderate to high risk of mortality compared with many other acute care diseases, even among patients without hypotension. Physicians should be aware of the poor prognosis of this group of patients, should monitor them closely, and give consideration to more aggressive treatment in order to prevent further progression to shock or death. Acknowledgments Dr Jones has received salary support from the National Institute of General Medical Sciences (R01 GM103799). Dr Puskarich received salary support from a Career Development Grant from the Emergency Medicine Foundation. Appendix 1. Representative search strategy for the Medline database b1950 to JULY 2013N 1. Lactates/(55420) 2. Acidosis, Lactic/(6720) 3. (((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections (1875648) 4. Systemic Inflammatory Response Syndrome (92052) 5. (Lactates) OR Acidosis, Lactic/(59597) 6. (((((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections)) OR Systemic Inflammatory Response Syndrome (1881924) 7. Prognosis/(11211486) 8. Outcome Assessment (Health Care)/(130252) 9. (Prognosis) OR Outcome Assessment (Health Care) (1186218) 10. (((((Lactates) OR Acidosis, Lactic/)) AND ((((((bacterial infections) OR brain abscess) OR central nervous system infections) OR infections)) OR Systemic Inflammatory Response Syndrome))) AND (((Prognosis) OR Outcome Assessment (Health Care)) (310) 11. Limit (#10) to HUMANS, AGE 19 + (193)
References [1] Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis in United States emergency departments. Crit Care Med 2007;35(8): 1928–36. [2] Angus D, Linde-Zwirble W, Lidicker J, Clermont J, Carcillo G, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29(7):1303–10. [3] Gaieski D, Edwards J, Kallan M, Carr B. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41(5): 1167–74. [4] Carlbom DJ, Rubenfeld GD. Barriers to implementing protocol-based sepsis resuscitation in the emergency department—results of a national survey. Crit Care Med 2007;35(11):2525–35. [5] Mikkelsen M, Gaieski D, Goyal M, Miltiades A, Munson J, Pines JM, et al. Factors associated with nonadherence to early goal-directed therapy in the ED. Chest 2010;138:551–8. [6] Paoletti F, Ripani U, Antonelli M, Nicoletta G. Central venous catheters: observations on the implantation technique and its complications. Minerva Anestesiol 2005;71(9):555–60. [7] Traore O, Liotier J, Souweine B. Prospective study of arterial and central venous catheter colonization and of arterial- and central venous catheterrelated bacteremia in intensive care units. Crit Care Med 2005;33(6): 1276–80. [8] Talmor D, Greenberg D, Howell MD, Lisbon A, Novack V, Shapiro N. The costs and cost-effectiveness of an integrated sepsis treatment protocol. Crit Care Med 2008;36(4):1168–74. [9] Jones AE, Troyer JL, Kline JA. Cost effectiveness of an emergency department based early sepsis resuscitation protocol. Crit Care Med 2011;39:1306–12. [10] Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent J-L. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996;171(2):221–6. [11] Jansen TC, van Bommel J, Bakker J. Blood lactate monitoring in critically ill patients: a systematic health technology assessment. Crit Care Med 2009;37(10): 2827–39. [12] Mikkelsen M, Miltiades A, Gaieski D, Goyal M, Fuchs B, Shah C, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med 2009;37:1670–7. [13] Puskarich M, Kline JA, Summers RL, Jones AE. Prognostic value of incremental lactate elevations in emergency department patients with suspected infection. Acad Emerg Med 2012;19(8):983–5. [14] Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goaldirected therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345(19):1368–677. [15] Dellinger R, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving Sepsis Campaign: international guidelines for the management of severe sepsis and septic shock 2012. Crit Care Med 2013;41(2):580–637. [16] Puskarich M, Trzeciak S, Shapiro N, Heffner AC, Kline JA, Jones AE, et al. Outcomes of patients undergoing early sepsis resuscitation for cryptic shock compared to overt shock. Resuscitation 2011;82(10):1289–93. [17] Cochrane handbook for systematic reviews of interventions 4.2.6 [updated September 2006]. In: Higgins JPT, Green S, editors. Cochrane Library. Issue 4 ed. Chichester, UK: John Wiley & Sons, Ltd.; 2006. [18] Liu V, Morehouse JJ, Soule J, Whippy A, Escobar GJ. Fluid volume, lactate values, and mortality in sepsis patients with intermediate lactate values. Ann Am Thorac Soc 2013;10(5):466–73. [19] Shapiro NI, Howell MD, Talmor D, Nathanson LA, Lisbon A, Wolfe RE, et al. Serum lactate as a predictor of mortality in emergency department patients with infection. Ann Emerg Med 2005;45(5):524–8. [20] Trzeciak S, Dellinger R, Chansky ME, Arnold R, Schorr C, Milcarek B, et al. Serum lactate as a predictor of mortality in patients with infection. Intensive Care Med 2007;33(6):970–7. [21] Howell MD, Donnino M, Clardy P, Talmor D, Shapiro NI. Occult hypoperfusion and mortality in patients with suspected infection. Intensive Care Med 2007;33(11): 1892–9. [22] Glickman SW, Cairns CB, Otero RM, Woods CW, Tsalik EL, Langley RJ, et al. Disease progression in hemodynamically stable patients presenting to the emergency department with sepsis. Acad Emerg Med 2010;17(4):383–90. [23] Song YH, Shin TG, Kang MJ, Sim MS, Jo IJ, Song KJ, et al. Predicting factors associated with clinical deterioration of sepsis patients with intermediate levels of serum lactate. Shock 2012;38(3):249–54. [24] Arnold RC, Sherwin R, Shapiro NI, O'Conner JL, Glaspey L, Singh S, et al. Multicenter obsevational study of the development of progressive organ dysfunction and therapeutic interventions in normotensive sepsis patients in the emergency department. Acad Emerg Med 2013;20(5):433–40. [25] Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31(4):1250–6. [26] Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, et al. Early lactate clearance is associated with improved outcome in severe sepsis and septic shock. Crit Care Med 2004;32(8):1637–42. [27] Puskarich M, Trzeciak S, Shapiro N, Arnold R, Heffner AC, Kline JA, et al. Prognostic value and agreement of achieving lactate clearance or central venous oxygen saturation goals during early sepsis resuscitation. Acad Emerg Med 2012;19(3): 252–8.
M.A. Puskarich et al. / Journal of Critical Care 29 (2014) 334–339 [28] Roe MT, Messenger JC, Weintraub WS, Cannon CP, Fonarow GC, Dai D, et al. Treatments, trends, and outcomes of acute myocardial infarction and percutaneous coronary intervention. J Am Coll Cardiol 2010;56(4):254–63. [29] Kucher N, Rossi E, De Rosa M, Goldhaber SZ. Massive pulmonary embolism. Circulation 2006;113:577–82. [30] Lichtman JH, Leifheit-Limson EC, Jones SB, Wang Y, Goldstein LB. 30-Day riskstandardized mortality and readmission rates after ischemic stroke in critical access hospitals. Stroke 2012;43:2741–7. [31] Marik PE, Cavallazzi R. Does the central venous pressure predict fluid responsivenses? An updated meta-analysis and a plea for some common sense. Crit Care Med 2013;41(7):1774–81. [32] Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34(6):1589–96.
339
[33] Puskarich M, Trzeciak S, Shapiro NI, Arnold RC, Horton JM, Studnek JR, et al. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011;39(9):2066–71. [34] Gaieski D, Mikkelsen M, Bank RA, Pines JM, Massone R, Furia FF, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergecy department. Crit Care Med 2010;38(4):1045–53. [35] Arnold RC, Shapiro NI, Jones AE, Schorr C, Pope J, Casner E, et al. Multi-center study of early lactate clearance as a determinant of survival in patients with presumed sepsis. Shock 2009;32(1):36–40. [36] Puskarich M, Trzeciak S, Shapiro NI, Albers AB, Heffner AC, Kliner JA, et al. Whole Blood Lactate Kinetics in Patients Undergoing Quantitative Resuscitation for Severe Sepsis and Septic Shock. Chest 2013;143(16):1548–53.
