Comparative outcome in patients with delirium tremens receiving care in emergency services only versus those receiving comprehensive inpatient care.

BACKGROUND: Delirium tremens (DT) is a medical emergency. Many cases are treated and discharged from emergency services (ES), after complete or partial resolution of delirium. Few receive comprehensive inpatient addiction treatment (CIAT) after the initial emergency management.
OBJECTIVE: The objective of this study was to compare 6-month outcomes of treatment in alcohol-dependence syndrome (ADS) patients presenting with DT receiving either only emergency care or emergency care along with CIAT.
MATERIALS AND METHODS: In this prospective observational study, all patients of ADS presenting in DT over a 1-year period were followed up for 6 months. Patients who received care only in the emergency services (ES) (111) were compared with patients who received ES followed by CIAT (90). Primary followup measure was regular followup (RFU) at outpatient department, and patients not presenting for RFU received telephonic followup (TFU). Alcohol use status was monitored at 6 months, as per Feuerlein and Küfner criteria.
RESULTS: Patients who received both ES and CIAT had better RFU compared to patients treated in the ES alone at 6 months (71/90 vs. 17/111, respectively, P < 0.005). CIAT also resulted in better combined follow-up (RFU and TFU) (85/90 vs. 60/111, respectively, P < 0.005). Compared to ES treatment group alone, ES plus CIAT group had fewer relapses (41/85 vs. 42/60, respectively, P < 0.05). The most common reason for direct discharge from ES was nonavailability of beds for inpatient treatment.
CONCLUSIONS: Merely emergency treatment of ADS patients presenting with DT does not provide satisfactory treatment outcome with respect to alcohol use. ES treatment followed by CIAT ensures better outcome in the form of fewer relapses and better follow up.

INTRODUCTION

Delirium tremens (DT) is a severe form of alcohol withdrawal that occurs in about 3%–5% of patients during withdrawal from alcohol.[1] DT usually occurs between 48 and 72 h after the last drink of alcohol and resolves 2–5 days after onset.[2] This condition is most commonly treated with benzodiazepines as well as corrective measures for dehydration, dyselectrolytemia, and vitamin deficiencies.[13] With timely medical management of this condition, mortality for DT has fallen from 35% to <5%.[3] However, in patients with underlying other systemic illnesses, the mortality rate of DT still continues to be still around 5%–15%.[4] Countries with increasing per capita consumption of alcohol have shown a rapid increase in the incidence of DT.[5]

With per capita alcohol consumption having increased from 1.6 L from 2003 to 2005 to 2.2 L from 2010 to 2012,[6] alcohol-related complications, including seizures and DT, are frequently encountered in clinical practice in India. Many patients presenting in DT are treated in emergency services (ES). Both the florid manifestations of DT and a preoccupation to rule out other causes of delirium lead to an exclusive focus on addressing recovery from the delirium. Once the patient recovers from DT, the treatment is construed to have been completed, and the underlying issues of alcohol dependence and the high risk for relapse to alcohol are not systematically addressed.[7] Further, many patients, since they are unable to recall their behavior during the episode of delirium, are neither able to appreciate the seriousness of the condition nor are adequately motivated for change.

Existing studies in alcohol dependence have compared outcomes in the treatment of alcohol-dependence syndrome (ADS) in general, between in- and out-patient settings.[8910] There have been no studies comparing different treatment approaches among persons presenting with DT. This study, therefore, attempted to compare long-term alcohol drinking outcomes in persons presenting with DT who received comprehensive inpatient addiction treatment (CIAT) to those who received only ES treatment.

MATERIALS AND METHODS

This prospective nonrandomized observational study was conducted at a tertiary care neuropsychiatry institute, as an auxiliary of a doctoral study, which was approved by the Institutional Review Board. Patients presenting to the psychiatry ES of the institute, between January 2013 and January 2014, with a presumptive diagnosis of DT as per the International Classification of Diseases 10 diagnostic criteria, were included in this study.[11] Such patients are generally managed by trainee psychiatrists under the supervision of qualified psychiatrists. Those who had a subsequent change of diagnosis based on clinical profile and supportive investigations (e.g., Wernicke's encephalopathy and delirium due to other causes), comorbid psychiatric illness, or severe cognitive damage were excluded from the study.

ES treatment is based on the current evidence-based therapeutic guidelines, which includes assisted detoxification with adequate benzodiazepines (choice of particular benzodiazepine depended on biochemical parameters of individual), multivitamin supplementation, correction of electrolyte and water imbalance, and other supportive measures.[1] After the delirium subsided, patients who gave informed consent were included in the study. Consenting patients were evaluated by a semi-structured data collection questionnaire to assess their sociodemographic details, alcohol consumption patterns as well as clinical profile. Thereafter, patients were either discharged from the ES due to a variety of reasons (described in the results) or were admitted for CIAT for 3–4 weeks. Patients discharged after ES treatment were prescribed tapering doses of benzodiazepines and multi-vitamins and were advised to undertake further treatment on an outpatient basis. CIAT consisted of forced abstinence in a protected environment, identification of factors responsible for maintaining dependence/relapse, motivational interviewing, individual sessions for relapse prevention, institution of appropriate anti-craving medications, appropriate psycho-social interventions, and group therapy sessions,[56] led by mental health professionals. Study participants were followed up till 6 months after discharge. The primary follow-up measure was regular follow up (RFU) at outpatient department, the rest were telephonic follow up (TFU). Informed consent is routinely obtained from patients at the time of registration for TFU.

Outcome (alcohol use status) at 6-month follow-up was evaluated using a time line follow back method. Outcome was classified according to the criteria of Feuerlein and Küfner.[12] Patients were judged to be abstinent (no subjective reports or objective indication of alcohol consumption), improved (no more than three drinking periods lasting for <1 week during the last 6 months [lapses] or <30 g [female] or <60 g [male] alcohol per day on a regular basis, no signs of pathological drinking, neither physical nor psychiatric disorders nor inpatient treatments due to alcohol consumption), or relapsed (more than three lapses or regular consumption of >30/60 g alcohol per day, alcohol-related disorders, or inpatient treatments during the last 6 months).

RFU in the present study refers to follow-up as advised by the treating team and includes patients who had kept at least the 1st, 3rd, and 6th months' contact. TFU refers to patients who were telephonically contacted to ascertain their status as they had not reported for hospital follow-up. Information was obtained from the patient and a significant other. Patients lost to follow-up (LFU) are those who did not maintain RFU and were not contactable through TFU (due to invalid number, wrong recipient, or no response despite three consecutive phone calls).

Statistical analysis was carried out using “R Commander Package” in R software, developed by R Core Team, version 3.1.0.[13] Descriptive statistics such as frequency, percentage, mean, and standard deviation (SD) were used to express the data. Chi-square test was used for comparison of categorical data between groups and appropriate parametric and nonparametric tests for continuous variables. All statistical tests were two tailed. P < 0.05 was considered to be statistically significant.

RESULTS

During the period of study, 278 patients with a presumptive diagnosis of DT reported to the ES, of whom sixty were excluded. These included patients with alcohol-induced psychotic disorder (twenty), other comorbid psychotic disorders (nine), bipolar affective disorders (nine), Wernicke's encephalopathy (seven), intoxication delirium (three), disulfiram-ethanol reaction presenting with delirium (one), and comorbid severe brain damage with neurocognitive deficits (11). The remaining 218 cases of DT were included in the study.

Baseline characteristics of the emergency services-only group

Of the 218 patients, 111 (51%) received only ES-based treatment. Mean age of patients in the ES-only group was 36.81 years (SD – 7.71). Seventy-eight (70.27%) patients in this group were early-onset-dependent drinkers (age of onset of ADS <25 years). Ninety (81.08%) patients had a positive family history of alcohol use disorders. Prior history of DT was present in 48 patients (43.24%). A majority (76 patients, 68.47%) had a rating of severe alcohol dependence as per the Severity of Alcohol Dependence Questionnaire (SADQ-C). Fifty-one patients (45.95%) had a current episode of withdrawal seizure. Average daily quantity of alcohol intake in the ES group was 19.54 units (SD – 6.42) [Table 1].

Table 1

Comparison of patients presenting in delirium tremens treated in emergency services only (ES) versus those receiving care in emergency services (ES) plus comprehensive inpatient care on selected sociodemographic and substance use variables

The common reasons for discharge of the 111 patients from ES included nonavailability of inpatient beds in 78 (70.27%), financial constraints in 25 (22.5%), and serious physical comorbidities warranting general hospital referral in 8 (7.2%) patients.

Mean duration of the hospital stay in ES-treated patients was 2.83 days (SD – 0.73).

Baseline characteristics of the emergency services + comprehensive inpatient addiction treatment group

Of the remaining 107 patients, ninety completed CIAT after receiving emergency care. Mean duration of inpatient stay was 24 days (SD – 5.8) among these patients. Average age of this group was 37.27 years (SD – 7.76). Sixty-eight patients (75.56%) had early-onset ADS. Seventy-five (83.3%) patients had a positive family history of alcohol use disorders. Thirty-nine (43.3%) patients had a previous history of DT. Mean daily alcohol consumption in this group was 19.84 (SD – 6.63).

Demographic and substance use variables between CIAT- and ES-treated patients did not show significant difference [Table 1].

Both groups were comparable in the detoxification received [Table 2]. Seventeen patients who did not complete CIAT and were discharged at request were excluded from the analysis.

Table 2

Detoxification in patients presenting with delirium tremens: comparison among patients treated in emergency services only (ES) and patients receiving care in emergency services plus (ES) comprehensive inpatient addiction treatment

While anti-craving medication was prescribed in all patients in the CIAT, only 17 patients in the ES-only group who came for RFU received anti-craving medication. The most common anti-craving drugs prescribed were baclofen (in 68.89% of the ES + CIAT group and 64.71% of the ES-only group); acamprosate in 20% and 23.53% and topiramate in 11.1% and 11.76% of the two groups, respectively. These differences were not statistically significant (P = 0.02).

Comparison of follow-up and outcome

The comparison of the rates of follow-up between the ES-only group and the combined group (ES + CIAT) is summarized in [Table 3].

Table 3

Comparison of follow-up rates among patients with delirium tremens who received emergency care only (ES) versus those receiving care in emergency services (ES) plus comprehensive inpatient addiction treatment

Of the ninety who completed CIAT, 71 patients (78.89%) were on RFU for 6 months. Fourteen (15.55%) were followed up by TFU and 5 (5.55%) were LFU. Of those followed-up cases (RFU + TFU = 85), 41 (48.24%) remained abstinent, 3 (3.53%) had improved, and 41 cases (48.24%) had relapsed into dependence pattern of drinking, at 6 months. Positive outcomes (improved and abstinent) were seen in 51.76% of patients (44/85). There was no mortality in this group.

Among the ES-only patients (n = 111), only 17 (15.32%) were RFU. Forty-three patients (38.73%) were followed up by TFU and 48 (43.24%) were LFU. Of those who could be followed up (RFU + TFU = 60), 42 patients (70%) had relapsed and 18 (15%) had remained abstinent at 6-month follow-up. Three patients had died during the 6-month follow-up period. The differences in regularity of follow-up and 6-month outcomes were significantly different between the CIAT and ES groups [Table 4].

Table 4

6-month outcome of patients admitted with delirium tremens in the two treatment groups (emergency care only [ES] and emergency treatment [ES] plus comprehensive inpatient addiction treatment)

Follow-up data among CIAT-completed cases (N = 90) at 1, 3, and 6 months showed a progressive reduction in abstinence and increase in relapses [Figure 1]. Similar analysis was not attempted with ED alone-treated cases, as the number of people who regularly followed up out of the total patients in the group was very small (17/111, 15.32%).

Figure 1

Treatment outcome in patients presenting with delirium tremens: In patients treatment in emergency servicesplus comprehensive inpatient care [ES+CIAT]. Figures in percentages

Figures 1 and 2 summarize ES + CIAT at baseline, 1, 3, and 6 months and ES alone at baseline and 6 months. The abstinent group includes patients maintaining abstinence and those who showed improvement as per the guidelines used. The relapsed group includes patients lost to follow up.

Figure 2

Treatment Outcome in patients presenting with delirium tremens treatment in the emergency services (ES only) Figures in percentages

Figure 1 illustrates the abstinence and relapse rates at baseline, 1, 3, and 6 months for the CIAT completers and Figure 2 illustrates these rates at baseline and 6 months. Due to the high LFU, the abstinent/relapsed rates were represented at baseline and at 6 months for ES-alone group. While the CIAT group showed significantly better outcome at 6 months compared to the ES-only group, there was a gradual decline in abstinence following discharge, indicating the need for intensive follow-up.

Comprehensive inpatient addiction treatment noncompleters

Of the 17 patients who did not complete CIAT, 11 (64.7%) were under RFU. Four patients (23.53%) received TFU and 2 (11.76%) were LFU. Among this group in whom 15 patients were contactable on follow-up, 6 (40%) were abstinent at 6 months and 9 (60%) relapsed into drinking. The rate of follow-up in this group which was admitted to CIAT but did not complete it was better than the group which was discharged after ES, although the results were not statistically significant. However, the small number in this group does not permit any generalization of the result [Table 5].

Table 5

6-month outcome of delirium tremens: comparison of patients who received emergency care only (ES) versus those receiving emergency treatment (ES) plus incomplete comprehensive inpatient addiction treatment

DISCUSSION

DT is an emergency medical condition, with untreated mortality up to 35%.[3] Substance use services must include adequate and effective management strategies for this severe complication of alcohol withdrawal. The observation of increased rates of DT noted in some countries[5] is generalizable to others where per capita consumption of alcohol has increased, including in India. Many patients presenting with DT are treated in the ES and are quickly discharged, in keeping with usual ES rapid turnover of patients. An important question that arises is whether it is appropriate to discharge such patients presenting with DT soon after resolution of delirium, without addressing the issues of their alcohol dependence. There is a temptation to do this, as it is often assumed that patients recovering from delirium may not be in a position to undergo motivation-enhancement and relapse prevention interventions.

DT usually manifests in patients with severe alcohol dependence. The present study highlights the importance of CIAT for this subset of patients with severe alcohol use disorders (AUDs) to ensure satisfactory treatment outcomes. The study also emphasizes the need for adequate inpatient treatment facilities in general hospitals for patients with DT. Similar conclusions were reached by Tiet et al., who recommend inpatient treatment for better results among those with severe substance use, although they do not specifically mention DT.[14]

This study compared treatment outcomes among patients with alcohol dependence presenting with DT who received one of the two treatment approaches. The first was treatment and discharge from the psychiatric ES, which perhaps represents standard treatment for DT in many treatment settings all over the world. The second is ES treatment followed by CIAT.

The study excluded patients with underlying neuropsychiatric, psychiatric, or severe medical comorbidities. Patients who were discharged from the ES settings would have had a resolution of delirium, but in view of the short duration of stay (mean: 2.83 days, SD – 0.73), would not have had an opportunity to be aware of the implication of the DT episode, and be engaged in counseling to motivate change and reduce relapse. In contrast, in the CIAT group, the inpatient stay allows for a greater and systematic engagement with the patient, including a detailed psycho-education about the episode of delirium as well as interventions to improve motivation and discuss strategies for relapse prevention.

The efficacy of inpatient treatment for alcohol dependence has been addressed in previous studies which indicated favorable (>40%) abstinence rate at 4-year follow-up.[15] An earlier study from our own center suggested that patients with severe alcohol dependence who opted for inpatient treatment did well.[10] However, we have not been able to locate any study on specific outcome of patients presenting with DT, although long-term follow-up is generally recommended for this condition. The findings of this study are thus significant to address this issue both in research and to guide service delivery.

Our results show a more favorable outcome in patients who received CIAT following the ES treatment. Relapses were significantly higher in the group treated with ES alone. RFU of 79% in those who underwent CIAT is an important finding of this study, as patients who present with DT are generally known to have very low rates of treatment adherence. Only 17% of ES-treated patients were followed up regularly. Those who were LFU among ES alone-treated group compared to CIAT group were significant (5.55% vs. 43.24%, respectively) [Table 2]. Even those who received CIAT, but did not complete it, did relatively better in terms of follow-up compared to the ES alone group [Table 5]. Three-month trend reflected 6-month outcome in terms of abstinence and relapse [Figures 1 and 2]. This has been reported by other studies (though not specifically for DT) and highlights the importance of early follow-up,[12] as it is well known that attrition is high in the longer term. Retention in follow-up improves outcome in tobacco dependence as well.[16]

Sixty of the 278 presumptive cases of DT (21.6%) were excluded due to other causes or presence of severe comorbid illnesses. This reflects the importance of careful evaluation for other causes of DT as well as comorbid psychiatric and physical conditions, in alcohol users presenting with delirium for good therapeutic outcome.

One hundred and eleven (51%) patients of the 218 confirmed DT cases were not admitted after emergency treatment. About 78 cases (70%) could not be admitted due to nonavailability of beds. This underscores the need for more inpatient care facilities to cater to the growing numbers of ADS patients presenting with DT. With AUD being a significant public health problem, the registration of 218 confirmed cases of DT during 1 year suggests that this is only the “tip of the ice berg.”

This study also shows that a large number of patients (48) were LFU in ES-treated group. This is an important finding of this study. Progressive increase in relapses with passage of time, in the early months posttreatment, calls for intensive after-care facilities to improve treatment retention. Retaining persons in treatment, on the other hand, is known to progressively improve treatment outcome.[16]

There are several limitations of this study, as this was an observational, nonrandomized study. The study could have been strengthened by the use of structured assessments for delirium, as well as for outcome assessments. Although the severity of delirium was not assessed using a structured instrument, SADQ-C, which is a measure of severity of alcohol dependence and could thus be a proxy measure for severity of withdrawal, was comparable in both groups. Although the major comorbidity was excluded, an in-depth comparison of minor comorbidities was not possible as this was an observational study. Other drawbacks are a lack of randomization and blinding, the confounding effects of pharmacotherapy (although detoxification regimens were comparable, anticraving medications were not prescribed to a majority in the ES group who failed to come for follow-up), patient-related variables such as the distance from the hospital, as well as the desirability of a follow-up >6 months. The reliability of TFU may be questionable. However, at our center, we have found self-report, corroborated by a family member, to be reliable. A cost-benefit analysis may help to support policy change in the approach to the treatment of patients presenting with DT.

Despite these limitations, the present study highlights the need for extended inpatient care of patients presenting with DT, which appears to be associated with significantly better treatment retention and outcome compared to brief emergency services treatment, which appears to be the current norm. Such a stay can aid cognitive recovery and provision of enhanced psychosocial support.

CONCLUSIONS

DT (alcohol withdrawal delirium) is a life-threatening emergency frequently treated by physicians, intensivists, or psychiatrists in ES. Many patients presenting with DT treated at ES are discharged from hospital without CIAT. This happens for a variety of reasons including nonavailability of beds and financial constraints. This study highlights the importance of CIAT along with emergency treatment of DT to significantly improve treatment outcomes. This study provides evidence for policy change to enhance and strengthen inpatient treatment facilities to provide CIAT for ADS patients presenting with DT. This group also needs intensive aftercare to improve treatment retention.

Such an isolated study on 6-month treatment outcome in DT patients was not found in literature. Hence, this study will serve as a scientific basis for further studies on DT in specific treatment locations.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.