Hemodynamic Response to Vasopressin Dosage of 0.03 units/min vs. 0.04 units/min in Patients with Septic Shock

Abstract

Background:

Arginine vasopressin (AVP) is suggested as an adjunct to norepinephrine in patients with septic shock. Guidelines recommend an AVP dosage up to 0.03 units/min, but 0.04 units/min is commonly used in practice based on initial studies. This study was designed to compare the incidence of hemodynamic response between initial fixed-dosage AVP 0.03 units/min and AVP 0.04 units/min.

Methods:

This retrospective, multi-hospital health system, cohort study included adult patients with septic shock receiving AVP as an adjunct to catecholamine vasopressors. Patients were excluded if they received an initial dosage other than 0.03 units/min or 0.04 units/min, or AVP was titrated within the first six hours of therapy. The primary outcome was hemodynamic response, defined as a mean arterial pressure ≥65 mm Hg and a decrease in catecholamine dosage at six hours after AVP initiation. Inverse probability of treatment weighting (IPTW) based on the propensity score for initial AVP dosage receipt was utilized to estimate adjusted exposure effects.

Results:

Of the 1536 patients included in the observed data, there was a nearly even split between initial AVP dosage of 0.03 units/min (n=842 [54.8%]) and 0.04 units/min (n=694 [45.2%]). Observed patients receiving AVP 0.03 units/min were more frequently treated at the main campus academic medical center (96.3% vs. 52.2%, p<0.01) and in a medical intensive care unit (87.4% vs. 39.8%, p<0.01). The IPTW analysis included 1379 patients with achievement of baseline covariate balance. There was no evidence for a difference between groups in the incidence of hemodynamic response (0.03 units/min 50.0% vs. 0.04 units/min 53.1%, adjusted relative risk 1.06 [95% CI 0.94, 1.20]).

Conclusions:

Initial AVP dosing varied by hospital and unit type. Although commonly used, an initial AVP dosage of 0.04 units/min was not associated with a higher incidence of early hemodynamic response to AVP in patients with septic shock.

Introduction

Septic shock mortality continues to be around 38% in North America,¹ therefore the therapeutic approach must be optimized. Surviving Sepsis Campaign Guidelines suggest arginine vasopressin (AVP) dosage up to 0.03 units/min as an adjunct to norepinephrine (NE) to increase mean arterial blood pressure (MAP) to goal or to decrease NE dosage in septic shock patients.² Due to the vasoconstrictive properties of AVP, the drug may be utilized similarly to other vasopressors with the dosage titrated to effect.^3–5 Indeed, the largest study of AVP titrated blinded AVP from 0.01 units/min to 0.03 units/min to target a goal MAP, which informed the dosing recommendation in the Surviving Sepsis Campaign Guidelines.⁵ However, because a relative vasopressin deficiency may exist in patients with septic shock, clinicians may add exogenous fixed-dosage AVP to catecholamine therapy as a form of endocrine replacement therapy.^{3, 4, 6} Initial studies of AVP for patients with septic shock used a fixed dosage of 0.04 units/min,^6–8 leading many clinicians to utilize this approach. In a survey of 200 critical care pharmacists, 87 respondents (44%) recommended an initial AVP dosage of 0.03 units/min without titration, 84 respondents (42%) recommended an initial AVP dosage of 0.04 units/min without titration, and 23 respondents (12%) recommended an initial AVP dosage ≤0.04 units/min with titration.⁹ Higher AVP dosages may lead to more pronounced decreases in NE dosage over time,¹⁰ but in some studies higher AVP dosages have been associated with a higher incidence of mortality, cardiac arrest, and adverse effects; highlighting the need to define the correct dose of AVP.^11–13 The optimal dosage of AVP for patients with septic shock has not been established and fixed AVP dosages of 0.03 units/min and 0.04 units/min have not been previously compared.^{3, 14} This study was designed primarily to compare the incidence of hemodynamic response in patients receiving fixed-dosage AVP 0.03 units/min vs. 0.04 units/min. Secondarily, the study sought to compare other patient-centered outcomes, vasoactive medication costs between AVP dosage regimens, and describe the prevalence of AVP dosage regimens in a large health-system.

Material and Methods

This retrospective, multi-hospital health system, observational cohort study evaluated patients admitted to a medical, surgical, or mixed medical/surgical intensive care unit (ICU) in one of eight hospitals in the Cleveland Clinic Health System between January 2012 and November 2017. Adult patients (≥18 years) were electronically screened for the presence of septic shock based on the Centers for Disease Control and Prevention Adult Sepsis Event definition.¹⁵ Specifically, patients were identified as having septic shock if they met each of the following criteria: blood cultures obtained, receipt of at least four days of broad-spectrum antibiotics (or until the time of their death, if sooner) initiated within two days of blood cultures, initiation of vasopressors within two days of blood cultures, and serum lactate concentration ≥2 mmol/L at the time of vasopressor initiation. Identified patients with septic shock initiated on AVP as an adjunct to a catecholamine vasopressor were included in the study. Those with documented initiation of a catecholamine vasopressor and AVP simultaneously were not included in order to avoid inclusion of patients previously started on AVP at another hospital and subsequently transferred to a study hospital. Patients initiated on an AVP dosage other than 0.03 units/min or 0.04 units/min and those in whom AVP was titrated within the first six hours of therapy were excluded. Only the first episode of exposure to AVP per patient was evaluated. Cohorts were developed based on initial AVP dosage of 0.03 units/min or 0.04 units/min.

Patients were treated in accord with the Surviving Sepsis Campaign guidelines available at the time, as directed by their primary care providers.^{2, 16} During the study time period the medical ICU at the main campus academic medical center had in place a vasoactive agent algorithm suggesting (but not requiring) an initial AVP dosage of 0.03 units/min without titration. However, other study ICUs did not have a vasoactive agent algorithm in place during the study time period and AVP dosage was left to provider discretion. The general approach to treating patients with septic shock, including timely antibiotic administration, adequate fluid resuscitation, and initial vasopressor use of norepinephrine did not change over the duration of the study.

The primary outcome measure was the incidence of hemodynamic response, defined as a decrease in NE-equivalent catecholamine dosage with a MAP ≥65 mm Hg at six hours after initiation of AVP. This definition of hemodynamic response was selected as the primary outcome because it has been independently associated with ICU mortality.¹⁷ Patients who died within six hours of AVP initiation were adjudicated as non-responders. Secondary outcomes included the prevalence of AVP dosage regimens (with accounting for patients excluded for alternative AVP dosage regimens), 28-day mortality, ICU length of stay, vasoactive medication duration overall and after AVP initiation, and total vasoactive medication costs during the shock episode. Total vasoactive medication costs were calculated based on pharmacy dispensing records of all vasoactive agents, with accounting for dispenses that were returned unused. In order to standardize comparisons due to rising drug costs over time, vasoactive wholesale acquisition cost (WAC) from November 2017 was used to calculate total vasoactive medication costs (e-Table 1 in the Supplemental Digital Content).¹⁸ Exposure time for 28-day mortality was defined as the time from AVP initiation until death within 28 days, with censoring of alive patients at 28 days. Tertiary outcomes included NE-equivalent catecholamine dosage change at six hours after AVP initiation, maximum NE-equivalent catecholamine dosage and lactate concentration in the 24 hours after AVP initiation, sequential organ failure assessment (SOFA) score change at 48 hours after AVP initiation, and total AVP duration. NE-equivalent catecholamine dosage change at six hours after AVP initiation and SOFA score change at 48 hours after AVP initiation were only calculated for patients surviving for the respective time durations. NE-equivalent catecholamine dosages were calculated as [NE (mcg/min)] + [Epinephrine (mcg/min)] + [Dopamine (mcg/kg/min)/2] + [Phenylephrine (mcg/min)/10].⁵ Broad spectrum antibiotic receipt was defined per the United States Centers for Medicare & Medicaid Services Early Management Bundle, Severe Sepsis/Septic Shock (SEP-1) quality measure.¹⁹ Acute kidney injury was defined per the Kidney Disease: Improving Global Outcomes Acute Kidney Injury definition.²⁰ Additional definitions are outlined in e-Table 2 in the Supplemental Material.

Patients in the dosage cohorts were likely to differ systematically due to confounding by indication, therefore measured confounding was controlled by using inverse probability of treatment weighting (IPTW) based on propensity scores.^{21, 22} The propensity score was developed based on variables thought to be clinically associated with initial AVP dosage receipt and the primary outcome (e-Table 2 in the Supplemental Digital Content).²¹ Patients with missing data could not have a propensity score calculated, therefore complete case analysis was implemented. Propensity scores were trimmed at the 5% margins to minimize areas of non-overlap in score distribution, prevent extreme weights, and reduce bias due to unmeasured confounders.^22–25 Propensity score distributions according to AVP dosage receipt are shown in e-Figure 1 in the Supplemental Digital Content. Observations were then weighted by the inverse probability of receiving an initial AVP dosage of 0.04 units/min, which created a pseudosample where the average treatment effect in the whole population could be evaluated. Baseline characteristics and outcomes are expressed as weighted means or medians for continuous variables based on distribution, and weighted percentage for categorical variables in the pseudosample, and were not directly observed values. Covariate balance in the observed data and weighted sample was assessed using absolute standardized differences, with values above 10% considered meaningful imbalance.²¹ Treatment effects using IPTW were estimated by fitting a generalized estimating equation model using a robust, “sandwich”-type covariance approach to account for within-subject correlation, presenting relative risk and mean (or percent) differences with 95% confidence intervals (CI) reported.^{21, 22, 26, 27} Continuous outcomes that did not satisfy the normal distribution assumption on initial analysis were log-transformed (which satisfied the normal distribution assumption) for analysis and transformed back for presentation, such that the mean difference indicated percent difference between dosage groups. Additionally, a Cox proportional hazards regression model was developed for 28-day mortality using a robust, “sandwich”-type covariance approach presenting hazard ratio with 95% CI reported. Categorical variables to assess the prevalence of AVP dosage regimens in the observed (non-weighted) population were analyzed with the chi-square test. Analyses were performed based on an overall significance level of 0.05, using SAS software (version 9.4, Cary, NC). This study was approved by the Cleveland Clinic Institutional Review Board (approval number 19–162) with a waiver of informed consent.

Results

Patient population and dosage prevalence

A total of 5810 patients with sepsis and an order for a vasoactive medication were screened; 3935 did not meet inclusion criteria and 339 were excluded (Figure 1). Exclusion from the study for an initial AVP dosage other than 0.03 units/min or 0.04 units/min (n=220 [11.7% of 1875 inclusion-eligible patients receiving AVP]) or early AVP dosage titration (n=119 [6.3% of 1875 inclusion-eligible patients receiving AVP]) was infrequent. Of the 1536 patients included in the observed (non-weighted) data, 842 (54.8%) received initial fixed-dosage AVP 0.03 units/min and 694 (45.2%) received initial fixed-dosage AVP 0.04 units/min. Observed patients receiving AVP 0.03 units/min were more frequently treated at the main campus academic medical center (n=811 [96.3%] vs. n=362 [52.2%], p<0.01) and in a medical ICU (n=736 [87.4%] vs. n=276 [39.8%], p<0.01). The majority of observed patients treated in a surgical ICU (n/N 187/226 [82.7%]) or mixed medical/surgical ICU (n/N 231/298 [77.5%]) received AVP 0.04 units/min. Additional baseline characteristics are presented in Table 1, with several meaningful differences between groups in the observed data. After propensity score estimation and trimming, 1379 patients were included in the IPTW analysis; 770 patients received AVP 0.03 units/min and 609 patients received AVP 0.04 units/min. Balance across all baseline covariates was achieved with weighting (Table 1, e-Figure 2 in the Supplemental Digital Content). Patients included in the IPTW analysis were severely ill with baseline weighted mean SOFA score of 14 and Acute Physiology and Chronic Health Evaluation III score of 107. At AVP initiation, patients were receiving a weighted mean NE-equivalent catecholamine dosage of 25 mcg/min with a lactate concentration of 3.6 mmol/L.

Figure 1.

Cohort build for initial vasopressin dosage 0.03 units/min vs. 0.04 units/min.

AVP = arginine vasopressin.

Table 1.

Characteristics of the Patients Before and After Inverse Probability of Treatment Weighting

	Observed Data (N = 1536)			Inverse Probability of Treatment-Weighted Data (N = 1379a, 2782b)
Characteristic	AVP 0.03 units/min (n = 842)	AVP 0.04 units/min (n = 694)	ASD (%)	AVP 0.03 units/min (n = 770a, 1397b)	AVP 0.04 units/min (n = 609a, 1385b)	ASD (%)
At Shock Onset
Age (years), mean	61.3	64.5	16.9	63.1	62.3	3.4
Male sex, %	51.0	52.3	2.0	51.8	52.1	0.6
Body weight (kg)	82.4	83.1	2.2	81.6	82.8	1.6
White/Caucasian race, %	68.5	73.3	6.3	72.2	70.5	3.7
Comorbid diseases, %
Cirrhosis	22.0	13.0	20.3	17.4	16.9	1.2
COPD	20.2	24.9	6.0	23.1	21.8	3.3
Diabetes	25.4	31.6	5.1	26.3	26.3	0.0
ESRD	12.5	5.6	17.3	8.5	8.5	0.0
Hepatic failure	18.1	11.0	18.6	15.0	15.1	0.4
Immune suppression	20.1	13.7	13.7	16.3	17.7	3.9
Leukemia, lymphoma, or myeloma	10.9	4.5	20.8	7.7	8.5	2.9
Medical ICU admission, %	87.4	39.8	107.4	66.4	67.1	1.5
SOFA score, mean	14.4	13.2	29.8	13.9	14.0	1.1
APACHE III score, mean	110.8	104.4	17.0	107.4	107.8	0.8
Mechanical ventilation, %	51.3	43.4	10.7	48.5	49.4	1.9
Broad spectrum antibiotic receipt, %	94.7	94.2	2.6	94.7	94.3	1.9
At AVP Initiation
Fluid bolus volume given (liters)	1.4	2.0	22.1	2.0	1.7	1.1
Fluid balance (liters positive)	2.8	3.2	0.9	2.8	3.2	2.2
NE as initial vasoactive medication, %	96.2	95.8	0.1	96.3	96.1	1.3
NE-equivalent catecholamine dosage (mcg/min)	26.0	20.0	15.4	25.0	25.0	1.2
MAP (mm Hg)	67.0	67.0	6.0	67.0	67.0	0.8
Lactate concentration (mmol/L)	3.6	3.6	5.6	3.5	3.6	1.6
Time elapsed from catecholamine start (hours)	6.5	5.1	10.8	6.4	5.8	4.3
AKI, %	59.4	59.8	3.8	60.3	60.6	0.5
During Shock Course
Hydrocortisone exposure, %	54.2	55.6	2.9	56.4	59.8	6.8

Data presented as median (unless otherwise specified) for observed data, and weighted statistics for inverse probability of treatment-weighted data.

^aActual sample size.

^bEffective sample size after inverse probability of treatment weighting.

AKI = acute kidney injury, APACHE = acute physiologic and chronic health evaluation, ASD = absolute standardized difference, AVP = arginine vasopressin, COPD = chronic obstructive pulmonary disease, ESRD = end-stage renal disease, ICU = intensive care unit, MAP = mean arterial blood pressure, NE = norepinephrine, SOFA = sequential organ failure assessment score.

Patient outcomes

There was no evidence to support a difference between AVP dosage groups in the incidence of hemodynamic response (0.03 units/min 50.0% vs. 0.04 units/min 53.1%, adjusted relative risk 1.06 [95% CI 0.94, 1.20]) (Table 2). Additionally, there was no evidence to support a difference between groups in 28-day mortality (60.9% vs. 55.9%, adjusted hazard ratio 0.87 [95% CI 0.74, 1.04]) (Figure 2) or ICU length of stay (5.8 days vs. 5.3 days, −7.8% difference [95% CI −19.4%, 5.5%]). Although the maximum NE-equivalent catecholamine dosage within 24 hours of AVP initiation was lower in the 0.04 units/min group (35.4 mcg/min vs. 31.8 mcg/min, −10.1% difference [95% CI −17.5%, −2.0%]), there was no difference detected between groups in vasoactive medication duration overall or vasoactive medication duration after AVP initiation (Table 2). The weighted mean total vasoactive medication cost per patient was $874 in the AVP 0.03 units/min group and $967 in the AVP 0.04 units/min group (difference 10.6% [95% CI −0.3%, 22.6%]).

Table 2.

Outcomes in the Study Cohort Before and After Inverse Probability of Treatment Weighting

	Unadjusted Values (N = 1536)			Adjusted Values (N = 1379a, 2782b)
Characteristic	AVP 0.03 units/min (n = 842)	AVP 0.04 units/min (n = 694)	RR, HR, or Difference, (95% CI)c	AVP 0.03 units/min (n = 770a, 1397b)	AVP 0.04 units/min (n = 609a, 1385b)	RR, HR, or Difference, (95% CI)c
Primary outcome
Hemodynamic response, %	51.1	51.9	1.02 (0.92, 1.13)d	50.0	53.1	1.06 (0.94, 1.20)d
Secondary outcomes
28-Day mortality, %	66.8	49.9	0.66 (0.58, 0.76)e	60.9	55.9	0.87 (0.74, 1.04)e
ICU length of stay (days)	6.0	5.1	−15.0% (−23.5%, −5.6%)f	5.8	5.3	−7.8% (−19.4%, 5.5%)f
Total vasoactive medication duration (days)	2.6	2.3	−10.7% (−18.3%, −2.5%)f	2.4	2.5	1.5% (−9.7%, 14.1%)f
Vasoactive medication duration after AVP initiation (days)	1.9	1.8	−2.4% (−11.2%, 7.4%)f	1.8	1.9	5.0% (−7.2%, 18.9%)f
Total vasoactive WAC ($)	924	906	−1.9% (−9.0%, 5.7%)f	874	967	10.6% (−0.3%, 22.6%)f
Tertiary outcomes
NE-equivalent catecholamine dosage change at 6 hours after AVP (mcg/min)	−0.1	−2.7	−2.6 (−5.5, 0.4)g	−0.7	−3.5	−2.8 (−6.1, 0.5)g
Maximum NE-equivalent catecholamine dosage within 24 hours after AVP (mcg/min)	35.6	29.7	−16.5% (−22.0%, −10.5%)f	35.4	31.8	−10.1% (−17.5%, −2.0%)f
Maximum lactate within 24 hours after AVP (mmol/L)	5.8	5.6	−2.3% (−8.2%, 3.9%)f	5.6	5.6	−0.1% (−7.5%, 8.0%)f
SOFA score change at 48 hours after AVP	−1.8	−1.8	0.1 (−0.3, 0.4)g	−1.9	−1.8	0.1 (−0.4, 0.6)g
AVP duration (days)	1.3	1.2	−5.7% (−14.3%, 3.8%)f	1.2	1.2	−1.6% (−13.3%, 11.6%)f

Data presented as mean (unless otherwise specified) for observed data, and weighted statistics for inverse probability of treatment-weighted data.

^aActual sample size.

^bEffective sample size after inverse probability of treatment weighting.

^cRelative risk, hazard ratio, and differences are for the AVP 0.04 units/min group relative to the AVP 0.03 units/min group.

^dRelative risk.

^eHazard ratio.

^fPercent difference. Outcome was log-transformed before analysis to satisfy the normal distribution assumption and transformed back for presentation.

^gMean difference.

AVP = arginine vasopressin, ICU = intensive care unit, NE = norepinephrine, RR = relative risk, SOFA = sequential organ failure assessment score, WAC = wholesale acquisition cost

Figure 2.

Weighted Kaplan-Meier Survival Curves for 28-Day Mortality of Patients Treated with Initial Vasopressin Dosage 0.03 units/ min vs. 0.04 units/min.

The dark blue line represents the AVP 0.03 units/min group, with 95% confidence intervals represented in the light blue shaded area. The dark red line represents the AVP 0.04 units/min group, with 95% confidence intervals represented in the light red shaded area. Weighted probability of survival was determined with inverse probability of treatment weighting. Numbers below the figure represent the number of patients at risk at each corresponding time point. The represented number of patients is based on the effective sample size after inverse probability of treatment weighting. Actual sample size in the weighted analysis were n = 770 in the AVP 0.03 units/min group and n = 609 in the AVP 0.04 units/min group. Adjusted HR 0.87 (95% CI 0.74, 1.04) for the AVP 0.04 units/min group. AVP = arginine vasopressin, CI = confidence interval, HR = hazard ratio.

Discussion

In this large retrospective, multi-hospital health system, observational cohort study we did not detect evidence for a difference in early hemodynamic response to AVP between initial fixed-dosage regimens of 0.03 units/min and 0.04 units/min. We observed a high prevalence of AVP 0.04 units/min usage, with differences in dosage utilization based on center and ICU type. We did not detect differences between groups in secondary outcomes, including mortality. The current study aligns with the Surviving Sepsis Campaign research priorities calling for studies evaluating vasoactive medication dosages,²⁸ and supports the 2016 Surviving Sepsis Campaign guideline recommendations for an AVP dosage up to 0.03 units/min.²

In a previous study evaluating factors associated with hemodynamic response to fixed-dosage AVP, there was no difference detected in initial AVP dosage between AVP responders and non-responders.¹⁷ In a subsequent study, no difference was detected in vasopressin plasma concentration between AVP hemodynamic responders and non-responders, even when corrected for exogenous AVP dosage administered.²⁹ Our findings in the current study are consistent with these studies where a higher AVP dosage was not associated with the incidence of early hemodynamic response to AVP. In contrast, a randomized controlled trial comparing AVP 0.033 units/min versus 0.067 units/min found lower NE dosage requirements over the 48 hour study period with the higher AVP dosage regimen.¹⁰ The differing findings between the current study and the randomized controlled trial are likely due to different definitions of the catecholamine-related outcomes after AVP initiation, or less divergent AVP dosage regimens in the current study. Future studies need to elucidate the optimal AVP dosage strategy for patients with septic shock.

Our observed prevalence of AVP dosage regimens in the current study is similar to the proportion of dosage regimen recommendations reported in a survey representative of critical care pharmacists across the United States.⁹ In both studies there was a near-equal split of initial fixed-dosage AVP 0.03 units/min and 0.04 units/min, and titration of AVP was relatively rare. As such, the AVP dosage regimen prevalence we observed in the current study may be similar to those utilized in practice across the United States. The difference in AVP dosage regimen utilization by hospital and ICU type likely indicates differences in unit-based culture for AVP dosage ordering. The rationale for these differences in unit-based ordering of AVP dosing should be further explored.

Initial AVP dosing can have important implications for vasoactive medication-related acquisition cost. Utilizing the AVP cost basis from the current study, the WAC for AVP 0.04 units/min was $438 per day and AVP 0.03 units/min was $329 per day; a difference of $109 per patient per day. In the current study, mean total vasoactive medication cost was 10.6% (95% CI −0.3%, 22.6%) higher with AVP 0.04 units/min despite lower early NE dosage requirements with this regimen, and no differences detected between groups in total vasoactive medication duration or AVP duration. This suggests that the higher total vasoactive medication cost observed with AVP 0.04 units/min was likely due to higher AVP dosage-specific cost with this regimen that was not offset by lower catecholamine-related costs. Recent data from a payer’s perspective indicated adjunctive AVP appears to be cost-effective when compared to other vasopressors for patients with septic shock regardless of the drug acquisition cost, but hospital pharmacy medication budgets are often separated from total costs of care.³⁰ Therefore, in light of the current findings of likely higher vasoactive medication costs with AVP 0.04 units/min without patient-centered outcome benefits over AVP 0.03 units/min, ICUs preferentially using fixed-dosage AVP 0.04 units/min should collaborate with their multi-professional team members to re-evaluate their AVP dosage practice.

In the current study we observed a lower maximum catecholamine dosage within 24 hours of AVP initiation in the AVP 0.04 units/min dosage group, but did not detect a difference between groups in NE-equivalent catecholamine dosage change at 6 hours after AVP initiation. Because fixed-dosage AVP reaches steady state concentrations within one to two hours after initiation, the observed difference in maximum catecholamine dosage within 24 hours of AVP initiation may have been due to center- or unit-specific maximum allowable catecholamine dosages. Importantly, the lower 24 hour maximum catecholamine dosage requirements in the AVP 0.04 units/min group did not translate to detected differences between groups in vasoactive medication duration or patient-centered outcomes.

Strengths of this study include its large sample size representing patients at multiple hospitals and ICU types, and rigorous design to control for confounding variables. The study is inherently limited by its observational nature with the inability to fully account for unmeasured confounders. Residual confounding is possible, but this risk is minimized with the use of IPTW with trimming of the propensity score margins.²⁴ Additionally, we did not specifically evaluate for adverse effects. However, even with AVP dosages up to 0.06 units/min, serious adverse effects were not more frequent with AVP than control.^{31, 32} Although digital ischemia is more frequent with AVP than control,^32–34 this adverse effect is inconsistently documented in the medical record. Consequently, it would have been unlikely for us to have detected a difference in adverse effects between AVP dosage regimens in the current study. Lastly, because we specifically designed this study to compare hemodynamic response between AVP fixed-dosage regimens of 0.03 units/min and 0.04 units/min, we excluded patients who had AVP dosage titration within the first six hours of therapy. Therefore, patients without a clinical response to their initial AVP dosage with an early increase in their AVP dosage, as well as those with a profound clinical response and an early AVP dosage decrease, were excluded. As such, although the definition we utilized aligned with previous literature, our reported incidence of hemodynamic response to AVP may be inaccurate. However, because study exclusion for AVP dosage titration was relatively infrequent (6.3%), we believe the degree of misclassification bias, if present, is minor. The impact of inclusion of AVP dosage titration actions in the definition of hemodynamic response to AVP should be further examined.

Conclusions

An initial fixed AVP dosage of 0.04 units/min, although commonly used, was not associated with a higher incidence of early hemodynamic response in patients with septic shock. Initial AVP dosing prevalence varied by hospital and ICU type. We did not detect evidence to support a difference in patient-centered outcomes between initial AVP fixed-dosage of 0.03 units/min or 0.04 units/min.