Analgesia and Sedation Requirements in Mechanically Ventilated Trauma Patients With Acute, Preinjury Use of Cocaine and/or Amphetamines

Bridgette Kram, PharmD; Shawn J. Kram, PharmD; Michelle L. Sharpe, PharmD; Michael L. James, MD; Maragatha Kuchibhatla, PhD; Mark L. Shapiro, MD


Anesth Analg. 2017;124(3):782-788. 

In This Article


Unexpectedly, the results of this study demonstrate that mechanically ventilated patients with a positive UDS for cocaine and/or amphetamines have widely variable daily opioid or sedative requirements that may not be different from patients with a negative UDS for these agents, when normalized for body weight. Cointoxication with alcohol and cannabinoids was common in this cohort, although it did not appear to affect requirements. Depth of sedation and clinical outcomes did not differ between groups. Stimulant-positive patients presenting with a TBI spent less time within the target RASS range and may have required intervention for elevated intracranial pressure more often, suggesting a more complicated neurological course.

Data examining the effect of drug use disorders and analgesic and sedative requirements during mechanical ventilation in the trauma or surgical patient populations are limited. De Wit et al[9] retrospectively examined sedative requirements in mechanically ventilated medical patients (n = 70) with a history of general substance use, including alcohol (31%), cocaine (7%), and other drugs within the past year, as identified by Diagnostic and Statistical Manual of Mental Disorders criteria. These patients with a history of a drug use disorder received more daily morphine equivalents (0.5 vs 0.1 mg/kg/day, P = .03) and lorazepam equivalents (0.5 vs 0.2 mg/kg/day, P = .004), whereas no difference was noted in propofol requirements, depth of sedation, and duration of mechanical ventilation. Not surprisingly, patient demographics in this study by de Wit et al[9] differed from our study population in several respects: they were older, had a history of alcohol or cocaine within the previous year, rather than acute use of cocaine and/or amphetamines, only a small proportion had a history of stimulant use specifically, and benzodiazepine administration was more common. In addition, the patients in our study received higher quantities of morphine equivalents than patients in this trial, which might be expected for a trauma cohort compared with a medical one; however, the 5-fold difference in opioid requirements was not reproduced in our study. We also did not appreciate any differences in the administration of adjunctive analgesics or sedative requirements.

Importantly, this study does not distinguish between drug users and abusers. Abuse may occur when a patient experiences adverse socioeconomic or health consequences that are related to the use of a substance, and it often leads to dependence and/or addiction.[19] However, this study is likely a more pragmatic approach because the ability to distinguish between the 2 is often not possible in trauma patients on presentation or early in their hospital course. In this study, 2 patients had a stimulant documented as a home medication. This finding may be conservative given that a large portion of trauma patients had unknown or incomplete past medical histories.

Although preinjury binge use of stimulants was less common, chronic stimulant use was documented for nearly one-half of stimulant-positive patients, and nearly 40% of patients had unknown use or they denied illicit substance use despite a positive UDS. Primary symptoms of stimulant withdrawal include agitation, depressed mood, fatigue, and discomfort. Although not observed in this study, it is possible that patients with chronic substance abuse may require higher doses of sedatives to mitigate this withdrawal syndrome. One potential explanation for this lack of difference may be because of the findings revealed in the subgroup analysis of patients with TBI. Although severity of injury scores at presentation and mean RASS were not statistically different, clinically, the stimulant-positive patients with TBI required intervention for elevated ICP more frequently and appeared to be more deeply sedated despite similar medication use. Less sedation may have been needed because of the evolution of the disease process and a more complicated neurologic course for the stimulant-positive patients, and subsequently, this may have influenced the lack of difference in sedation requirements observed in the larger cohort. These statements are made with caution because of the small sample size of the TBI subgroup.

One interesting finding in this study was the trend toward a greater use of ketamine in the stimulant-positive group. Ketamine administration may reduce perioperative opioid requirements and sensitize opioid receptors, thus it offers an attractive adjunctive agent in the trauma and surgical patient population.[20–23] Although our findings may reflect an institutional practice among the anesthesiology intensivists at our institution, the role of ketamine and other adjunctive analgesics for patients with a history of stimulant use warrants further exploration.

This study has several limitations. First, this was a retrospective study that, among other implications, suggests the potential for treatment bias regarding whether a clinician was aware of the drug screen results before drug selection and titration. There is also the potential for differing sedation practices from 2009 to 2013, with consensus guidelines for pain, agitation, and delirium recommending propofol and dexmedetomidine as first-line sedatives, while minimizing benzodiazepine use.[24] Yet, it is unlikely that these guidelines had a large effect on the results because they were only made available at the latter end of the study period. In light of emerging literature, a cultural transition toward targeting lighter levels of sedation may have occurred in the study period; however, this is difficult to characterize retrospectively, and included patients were equally distributed over time, so we expect that patients were sedated similarly. Drug shortages may have also affected the drug selection during this time period. For example, dexmedetomidine may have been used during a propofol shortage. Finally, registry data from 2006 to 2010 at our institution indicate that approximately 30% of patients receive a UDS at the time of admission after traumatic injury, and that women and patients ≥45 years of age are less likely to receive a UDS.[25] Although similar to previously published reports,[26] this biased screening process may have limited our sample size.

As a strength of its design, this study included patients in the stimulant-positive group based on positive results for cocaine and/or amphetamines and into the stimulant-negative group based on negative results for these agents, regardless of other positive results. Therefore, cointoxications were common in this study. Although this could have altered results, it is likely encountered in practice because cointoxication is common in this patient population.[27] This study is a realistic representation of sedation practice in this patient population over a period of time and offers us guidance on optimal sedation management of these patients.

In summary, achieving desirable pain control and depth of sedation may not require higher doses of analgesic and sedative medications for trauma patients presenting with acute intoxication of cocaine and/or amphetamines compared with their stimulant-negative counterparts. These findings should be interpreted cautiously in light of the wide CI observed in the propensity score-adjusted model. The findings, although unexpected, indicate that empirically increasing analgesic and sedative doses based on positive UDS results for these stimulants may not be necessary. Using a multimodal approach in the management of pain and agitation in this patient population, particularly in the role of ketamine, warrants further exploration.