Using a retrospective chart review, the authors compare the effectiveness of these two methods of administration in intensive care unit patients with septic shock.
Keywords: steroids, septic shock, hydrocortisone, continuous infusion, intermittent boluses, mortality benefit
Abstract
Purpose
To compare the effectiveness of continuous infusion of hydrocortisone versus intermittent boluses in the resolution of septic shock.
Methods
A retrospective chart review was performed to investigate the effects of low-dose hydrocortisone continuous infusion (200 mg per day) versus intermittent boluses (50 mg every six hours) in septic shock patients who did not respond to fluid resuscitation and vasopressors. The primary outcome was time to resolution of shock, defined by time from the initiation of hydrocortisone to time of vasopressor withdrawal when mean arterial pressure was greater than 65 mm Hg. Hospital length of stay, intensive care unit (ICU) length of stay, 28-day all-cause in-hospital mortality, and hyperglycemia were secondary outcomes.
Results
Of the 51 patients in the study, 33 received intermittent boluses of hydrocortisone. The median time to resolution of shock in the continuous group was three days, compared with a mean time of two days in the bolus group (P = 0.41). Mean hospital length of stay was 18.7 ± 19.4 days in the continuous group versus 18.6 ± 17.4 days in the bolus group (P = 0.77). Mean ICU length of stay was 10.4 ± 12.5 days in the continuous group versus 11.1 ± 9.2 days in the bolus group (P = 0.31). Thirteen patients (72%) in the continuous group died within 28 days, compared with 20 (60%) in the bolus group (P = 0.76). Five patients (28%) had hyperglycemic episodes after being given hydrocortisone in the continuous infusion group, compared with 20 patients (60%) in the bolus group (P = 0.04).
Conclusion
There was no significant difference in time to resolution of septic shock between continuous infusion (200 mg per day) and intermittent boluses (50 mg every six hours) of hydrocortisone. There were also no statistically significant differences in overall hospital length of stay, ICU length of stay, and 28-day all-cause in-hospital mortality. However, there was a significant difference in the incidence of hyperglycemia between the two groups, with patients in the bolus group experiencing more hyperglycemia than those in the continuous infusion group.
INTRODUCTION
Septic shock is the most serious manifestation of severe sepsis, which has an in-hospital mortality rate of 28%.1 Corticosteroids have a role as adjunctive therapy in septic shock through their anti-inflammatory effect of terminating systemic and tissue inflammation.2 The use of bolus hydrocortisone has been shown to be associated with a 10% absolute reduction in 28-day mortality.3 In addition, corticosteroids have also demonstrated significant improvement in cessation of vasopressor support compared with placebo.4 Although this mortality benefit was later called into question by Sprung and colleagues,5 their randomized, placebo-controlled trial concluded that hydrocortisone hastened the reversal of shock more quickly than placebo. As a result, corticosteroids have been recommended to restore cardiovascular homeostasis only when there is compelling evidence to do so—that is, when patients remain persistently hypotensive despite fluid resuscitation and vasopressor therapy.6
However, the optimal method of administering hydrocortisone remains unclear. The current Surviving Sepsis Campaign guidelines recommend continuous infusion over intermittent boluses to avoid hyperglycemia, but this is a weak recommendation (grade 2D).6 In a study by van Hooijdonk et al.,7 bolus hydrocortisone was associated with more fluctuation in glucose levels than placebo in critically ill patients, although the study did not evaluate continuous infusion of hydrocortisone. Loisa et al. showed that strict normoglycemia is more easily achieved if the hydrocortisone therapy is given to septic shock patients by continuous infusion. This approach also reduces the nursing workload needed to maintain tight blood glucose control.8 In Weber-Carstens et al., patients who were receiving continuous infusions of 200 mg per day of hydrocortisone at baseline were adjusted to a single bolus of 50 mg; mean blood glucose increased significantly from a baseline of 128 mg/dL to 154 mg/dL within six hours.9
METHODS
This study took place at Winthrop-University Hospital, a 591-bed acute-care teaching hospital in Mineola, New York, that has the following intensive care units (ICUs): medical, surgical, cardiac, neuroscience, and pediatrics. The study was approved by the hospital’s investigational review board. This retrospective chart review was designed to evaluate whether continuous infusion or intermittent boluses of hydrocortisone would show better results in patients with septic shock. We evaluated adults admitted to the surgical, medical, cardiac, and neuroscience ICUs with an ICD-9 code of septic shock who received both vasopressors and hydrocortisone between August 2014 and April 2016. Only patients who received either daily continuous infusion of hydrocortisone 200 mg or intermittent boluses of 50 mg every six hours were included in this study. Patients who received other variations of the hydrocortisone dosing scheme were not included. Patients were excluded if they had a documented hypersensitivity to steroid therapy; did not have a diagnosis of septic shock; previously received maintenance steroid therapy prior to shock; or received both modes of administration of hydrocortisone. The primary outcome was the time to resolution of shock defined by a mean arterial pressure (MAP) greater than 65 mm Hg measured from the time (in days) that patients were started on hydrocortisone until vasopressors were discontinued. Overall hospital length of stay, ICU length of stay, and 28-day all-cause in-hospital mortality were secondary outcomes. Twenty-eight-day all-cause in-hospital mortality was defined as the time that shock began until in-hospital patient death. Patients were considered to have a hyperglycemic episode if they had an average point-of-care glucose level greater than 155 mg/dL.10
STATISTICAL ANALYSIS
All statistical tests were two-sided. Continuous data were presented as means plus or minus standard deviations and the categorical data as proportions. The main endpoint was time to resolution of septic shock. Patient characteristics were compared between the groups (bolus versus continuous infusion) using the nonparametric Wilcoxon rank-sum test for continuous variables and Fisher’s exact test for categorical variables. In addition, a nonparametric log-rank test was used to compare the survival curves. The notes on Table 1 and Table 2 offer details.
Table 1.
Demographics and Clinical Characteristics
| Variables | Hydrocortisone Administration | P Values* | |
|---|---|---|---|
| Bolus (n = 33) | Continuous Infusion (n = 18) | ||
| Age, years | |||
| Mean ± SD | 69.8 ± 16.8 | 70.7 ± 21.0 | 0.51 |
| Median (IQR) | 76.5 (61–85) | 70 (59–82) | |
| Gender | |||
| Female, n (%) | 13 (39) | 6 (33) | 0.77 |
| Race, n (%) | |||
| African-American | 6 (18) | 3 (17) | |
| White | 25 (76) | 14 (78) | |
| Other | 2 (6) | 2 (6) | |
| Comorbidities, n (%) | |||
| Diabetes | 7 (21) | 2 (11) | 0.46 |
| Cardiovascular | 21 (64) | 9 (50) | 0.39 |
| Cancer | 9 (27) | 4 (22) | 0.75 |
| Hyperlipidemia | 5 (15) | 1 (6) | 0.41 |
P values from Wilcoxon rank-sum test for continuous variable (age) and Fisher’s exact test for all categorical variables.
IQR = interquartile range; SD = standard deviation.
Table 2.
Primary and Secondary Endpoint Comparisons Between Bolus and Continuous Infusion Groups
| Variables | Hydrocortisone Administration | P Values* | |
|---|---|---|---|
| Bolus (n = 33) | Continuous Infusion (n = 18) | ||
| Median time to resolution of shock, days | 2 | 3 | 0.41 |
| Hospital length of stay, days | |||
| Mean ± SD | 18.6 ± 17.4 | 18.7 ± 19.4 | 0.77 |
| Median (IQR) | 13 (7–22) | 13.5 (4–22) | |
| ICU length of stay, days | |||
| Mean ± SD | 11.1 ± 9.2 | — | 0.31 |
| Median (IQR) | 8 (4–14) | — | |
| 28-day all-cause in-hospital mortality, n (%) | 21 (64) | 13 (72) | 0.76 |
| Hyperglycemia, n (%) | 20 (60) | 5 (28) | 0.04 |
P values are from log-rank test for time to resolution variable, Wilcoxon rank-sum test for other continuous variables, and Fisher’s exact test for categorical variables.
ICU = intensive care unit; IQR = interquartile range; SD = standard deviation.
Survival estimates and cumulative event rates were compared with the Kaplan-Meier method using the time-to-event approach. The log-rank test was used to compare the Kaplan-Meier survival curves between the bolus and continuous infusion groups. Secondary outcomes were compared using the Wilcoxon rank-sum test and Fisher’s exact test as appropriate. SAS 9.4 was used for all the computations, and a P value of less than 0.05 was considered statistically significant.
RESULTS
Baseline Characteristics
From August 2014 through April 2016, 76 patients developed septic shock in the hospital. Of those, 25 patients were excluded due to age (younger than 18 years, n = 3), maintenance steroid therapy (n = 8), and/or receiving both bolus administration and continuous infusion of hydrocortisone (n = 14). The remaining 51 patients eligible for analysis consisted of 33 who received intermittent boluses (50 mg every six hours) and 18 who received continuous infusion (200 mg over 24 hours) of corticosteroid. The median age was 70 years in the continuous infusion group and 76.5 years in the bolus group. Seven patients (21%) in the bolus group had diabetes, compared with two (11%) in the continuous infusion group (P = 0.46). In addition, a higher proportion of patients had pre-existing cardiovascular diseases in the bolus group compared with the continuous infusion group (64% versus 50%; P = 0.39) (Table 1).
Outcomes
All patients were managed appropriately according to the Surviving Sepsis Campaign’s 2012 International Guidelines for Management of Severe Sepsis and Septic Shock in regard to fluid resuscitation and vasopressor usage.6 Initial fluid challenge included 30 mL/kg of crystalloids when necessary. Norepinephrine was the first-choice vasopressor, followed by epinephrine and/or vasopressin. No dopamine was used. Phenylephrine was used in three patients who had persistent hypotension and tachycardia despite adequate fluid resuscitation and who did not tolerate norepinephrine or epinephrine because of severe tachycardia. Vasopressor dosage was not part of the data collection.
Fifteen patients (45%) achieved shock resolution in the hydrocortisone bolus group versus five patients (28%) in the continuous infusion group. The median time to resolution of shock in the continuous infusion group was three days, compared with two days in the bolus group. Resolution of shock is defined as MAP greater than or equal to 65 mm Hg without vasopressor usage, given the adequate fluid resuscitation. There was no statistically significant difference in this outcome (log-rank P = 0.41). Figure 1 depicts the comparison of the time-to-event curves between the groups.
Figure 1.
Comparisons of Time to Resolution of Septic Shock Between Bolus and Continuous Infusion Groups
In regard to secondary outcomes, the median hospital length of stay and interquartile range (IQR) in patients receiving continuous infusion was 13 days (range, 7–22 days), compared with 13.5 days (range, 4–22 days) in the bolus group (P = 0.77). The median and IQR ICU length of stay in the continuous infusion group was eight days (range, 4–14 days), whereas patients in the bolus group stayed in the ICU for a median of six days (range, 2–14 days) (P = 0.31). Thirteen patients (72%) in the continuous group and 21 patients (64%) in the bolus group died in the hospital within 28 days of the onset of shock (P = 0.76). Twenty patients (60%) in the bolus group had a hyperglycemic episode after hydrocortisone administration, compared with five patients (28%) in the continuous infusion group (P = 0.04). Details appear in Table 2.
DISCUSSION
To date, there has been no study investigating whether one method of administering hydrocortisone shows better outcomes in mortality and hospital length of stay in septic shock. We assessed and compared the two methods in terms of these outcomes. There was no statistically significant difference in time to resolution of shock between patients who received either hydrocortisone continuous infusion or intermittent boluses in our study. There was also no significant difference in the other outcomes, such as overall hospital length of stay, ICU length of stay, and 28-day all-cause in-hospital mortality.
Our results showed a difference in the number of hyperglycemic episodes between the two groups. Previous studies have also demonstrated a consistent result in regard to hyperglycemia between the two methods of administration.8,9 In the absence of comparative studies of overall hospital outcomes, the Surviving Sepsis Campaign guidelines make only a low-strength recommendation of preferring continuous infusion of hydrocortisone to avoid hyperglycemia. Of note, our study utilized point-of-care (POC) glucose monitoring. The accuracy of the POC glucose monitor depends on device methodology and sample characteristics (e.g., variations in pH, blood oxygen, hematocrit, changes in microcirculation, and vasopressor therapy). These factors could significantly impact POC glucose-reading accuracy.11
LIMITATIONS
The study has several limitations. First, it is a retrospective cohort study. Second, the study-group sizes were not well matched. There were more diabetic patients in the bolus group, which could impact the incidence of hyperglycemia after hydrocortisone therapy. Based on the results of our study, larger trials with a prospective design are needed to evaluate the effects of the two methods of administration in septic shock. For now, clinicians should follow the recommendation from the current Surviving Sepsis Campaign guideline that favors continuous infusion over intermittent boluses of hydrocortisone, especially in patients with diabetes.
CONCLUSION
In this study, neither bolus nor continuous infusion of hydrocortisone demonstrated a statistically significant mortality benefit. Patients who received bolus hydrocortisone had significantly more episodes of hyperglycemia than those who received continuous infusion.
Footnotes
Disclosures: The authors report no commercial or financial interests in regard to this article.
REFERENCES
- 1.Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29(7):1303–1310. doi: 10.1097/00003246-200107000-00002. [DOI] [PubMed] [Google Scholar]
- 2.Annane D. Corticosteroids for severe sepsis: an evidence-based guide for physicians. Ann Intensive Care. 2011;1(1):7. doi: 10.1186/2110-5820-1-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Oppert M, Schindler R, Husung C, et al. Low-dose hydrocortisone improves shock reversal and reduces cytokine levels in early hyperdynamic septic shock. Crit Care Med. 2005;33(11):2457–2464. doi: 10.1097/01.ccm.0000186370.78639.23. [DOI] [PubMed] [Google Scholar]
- 4.Annane D, Sébille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288(7):862–871. doi: 10.1001/jama.288.7.862. Erratum in: JAMA 2008;300(14):1652. [DOI] [PubMed] [Google Scholar]
- 5.Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358(2):111–124. doi: 10.1056/NEJMoa071366. [DOI] [PubMed] [Google Scholar]
- 6.Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637. doi: 10.1097/CCM.0b013e31827e83af. [DOI] [PubMed] [Google Scholar]
- 7.van Hooijdonk RT, Binnekade JM, Bos LD, et al. Associations between bolus infusion of hydrocortisone, glycemic variability, and insulin infusion rate variability in critically Ill patients under moderate glycemic control. Ann Intensive Care. 2015;5(1):34. doi: 10.1186/s13613-015-0077-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Loisa P, Parviainen I, Tenhunen J, et al. Effect of mode of hydrocortisone administration on glycemic control in patients with septic shock: a prospective randomized trial. Crit Care. 2007;11(1):R21. doi: 10.1186/cc5696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Weber-Carstens S, Deja M, Bercker S, et al. Impact of bolus application of low-dose hydrocortisone on glycemic control in septic shock patients. Intensive Care Med. 2007;33(4):730–733. doi: 10.1007/s00134-007-0540-3. [DOI] [PubMed] [Google Scholar]
- 10.Keh D, Trips E, Marx G, et al. Effect of hydrocortisone on development of shock among patients with severe sepsis: the HYPRESS randomized clinical trial. JAMA. 2016;316(17):1775–1785. doi: 10.1001/jama.2016.14799. [DOI] [PubMed] [Google Scholar]
- 11.Klonoff DC. Point-of-care blood glucose meter accuracy in the hospital setting. Diabetes Spectr. 2014;27(3):174–179. doi: 10.2337/diaspect.27.3.174. [DOI] [PMC free article] [PubMed] [Google Scholar]