(31) An Open Label Pilot RCT of Low- vs High-Dose Telemedicine Buprenorphine Initiation

Background & Introduction: In the fentanyl era, buprenorphine initiation has become increasingly challenging. Fentanyl is highly lipophilic and can accumulate in adipose tissue and persist in the body even after withdrawal symptoms appear. Therefore, visible withdrawal may not reliably reflect receptor availability, increasing the risks for precipitated withdrawal and decreased adherence. To address these challenges, two divergent strategies have gained traction: high-dose (rapid receptor occupancy) and low-dose (gradual cross-taper) initiation. Despite clinical adoption, there have been no randomized controlled trials to our knowledge directly comparing these approaches among patients who use fentanyl.

Objective: To evaluate the feasibility and effectiveness of low- versus high-dose telemedicine buprenorphine initiation strategies in patients who use fentanyl.

Methods: This open-label pilot, randomized controlled trial enrolled adults with moderate/severe OUD and confirmed fentanyl use (urine toxicology) at an outpatient treatment program in Maryland (June-November 2024). Participants were randomized 1:1 to receive a 7-day buprenorphine initiation protocol using either a low-dose (0.5-16mg) or high-dose (2-24mg) schedule, each consisting of multiple daily doses. Dosing was observed via once-daily telemedicine sessions, followed by transition to office-based buprenorphine treatment. Following clinical guidelines, opioid withdrawal symptoms were treated with adjunctive medications (e.g., hydroxyzine, clonidine), which were assessed daily beginning Day 2. The primary outcome was treatment adherence, defined as completing the 7-day protocol without two consecutive missed observed doses. Secondary outcomes included daily changes in withdrawal symptoms (Subjective Opiate Withdrawal Scale; SOWS), precipitated withdrawal (SOWS increase ≥10 within three hours of dosing), and adjunctive medication use. Treatment adherence analyses followed intention-to-treat (ITT) and modified ITT (mITT; received ≥1 dose) approaches. Withdrawal trajectories were examined using mixed-effects regression models with random intercepts and slopes, adjusted for demographic and clinical covariates. Rates of precipitated withdrawal were compared using Fisher’s exact test and adjunctive medications use with mixed-effects linear regression. Participants were followed at one- and three-months post-initiation. The study was approved by WCG IRB.

Results: Thirty-nine participants were randomized (high-dose = 21; low-dose = 18). Four high-dose participants (19%) never initiated treatment. Reasons included fear of precipitated withdrawal (n=1), inability to reach withdrawal threshold (COWS ≥6; n=1), and loss to follow-up (n=2). In addition, three participants discontinued after initiating treatment (LD n=2, HD n=1) due to severe withdrawal symptoms (LD n = 1), loss to follow-up (LD n=1), and a participant-initiated unsupervised high-dose initiation that led to severe vomiting (HD n=1). Adherence: In the ITT sample, 71.4% of high-dose and 83.3% of low-dose participants met adherence criteria (p = 0.38). In the mITT sample, adherence was 88.2% (HD) versus 83.3% (LD; p = 0.68). Among study completers, adherence was 93.8% in both arms. Withdrawal trajectories: Among completers (n = 32; 210 observations), mean SOWS scores declined significantly over the 7-day period. Relative to Day 1, overall symptoms were lower by Day 5 (β = -6.5, p = 0.02), Day 6 (β = -8.3, p = 0.009), and Day 7 (β = -13.8, p < 0.001). There were no baseline differences between arms (p = 0.72). However, significant day × arm interactions were observed from Days 2-4 (ps < 0.03), suggesting that participants in the high-dose group experienced faster early symptom relief relative to those in the low-dose arm. A complementary model treating time as continuous showed consistent overall decline (β = -2.4, p < 0.001), supporting robustness of the categorical-day findings. Precipitated withdrawal: Precipitated withdrawal was assessed using standardized pre/post-dose SOWS measures, occurring in 20% of participants who initiated treatment, with similar rates observed between low-dose (22.2%) and high-dose (17.7%) arms (p = 0.73). No severe cases (SOWS increase ≥20) occurred. The single moderate event (SOWS increase ≥15) arose on day 6 in the high-dose arm but resolved with continued dosing. Adjunctive medication use: From days 2-7, participants used an average of five adjunctive medications per day. Adjunctive medication use trajectories differed significantly by treatment arm (day×arm interaction: p = 0.001). Medication use in the high-dose arm declined from 6.1 to 3.3 medications per day (days 2-7), while medication use in the low-dose arm remained relatively stable (5.3 to 5.6 medications/day, from days 2-7).

Conclusion & Discussion: While treatment adherence did not differ significantly between arms, distinct patterns emerged. High-dose participants who initiated treatment experienced faster early symptom reduction (Days 2-4) and required fewer adjunctive medications as the week progressed. These pilot results suggest that high-dose protocols may provide better withdrawal symptom control among patients who can successfully engage and tolerate this approach. However, 19% of high-dose participants never initiated treatment versus 0% in the low-dose arm, suggesting differential barriers to engagement (e.g., fear of precipitated withdrawal). Precipitated withdrawal occurred in roughly one-fifth of participants but was mild to moderate in all cases. While the study was not powered to detect meaningful differences, these pilot findings provide important preliminary estimates regarding the implementation of low- vs. high-dose buprenorphine initiation schedules via telemedicine, as well as the feasibility of study procedures. Longer-term outcomes at one- and three-months are being analyzed. Larger, adequately powered trials are needed to establish comparative effectiveness and refine best-practice buprenorphine initiation strategies in the era of fentanyl.

References:
1. Chambers LC, Hallowell BD, Zullo AR, et al. Buprenorphine dose and time to discontinuation among patients with opioid use disorder in the era of fentanyl. JAMA Netw Open. 2023;6(9):e2334540. doi:10.1001/jamanetworkopen.2023.34540

2. Shearer D, Young S, Fairbairn N, Brar R. Challenges with buprenorphine inductions in the context of the fentanyl overdose crisis: A case series. Drug and Alcohol Review. 2022;41(2):444-448. doi:10.1111/dar.13394

3. Tsui JI, Bhatraju EP, Hallgren KA, Rhew IC, Samet JH. High-dose buprenorphine treatment in the setting of fentanyl. JAMA Netw Open. 2025;8(8):e2528119. doi:10.1001/jamanetworkopen.2025.28119

4. Varshneya NB, Thakrar AP, Lambert E, Huhn AS. Opioid use disorder treatment in the fentanyl era. J Addict Med. 2023;17(1):118-119. doi:10.1097/ADM.0000000000001013

5. Suen LW, Chiang AY, Jones BLH, et al. Outpatient low-dose initiation of buprenorphine for people using fentanyl. JAMA Netw Open. 2025;8(1):e2456253. doi:10.1001/jamanetworkopen.2024.56253