A Tale of Two Cohorts: The Trials and Tribulations of Ever-Changing Orthopedic / Acute Pain Medicine Hospital Practices During the Execution of Military-Funded Clinical–Translational Research (2013–2021)

The implementation of the clinical trial reported in the accompanying article by Williams et al. [1] encountered several complications related to regulatory processes and the changing practices of the hospital at which the study was conducted. The present editorial reports on the complexities faced during trial execution, with the hope that this information may be helpful to other clinical researchers. In 2013, when we applied for funding for this project, single-injection nerve blocks with plain bupivacaine (BPV) were known to provide 15–20 hours of pain relief. Our institution, however, had used multimodal perineural analgesic with buprenorphine, clonidine, and dexamethasone (BCD) for more than 1,000 patients since 2011 [2]. These nerve blocks seemed to provide analgesia of a longer duration, but the method had not been compared against a known benchmark of plain BPV in a clinical trial. The military motivation for our research was to investigate whether single-injection multimodal perineural BPV-BCD nerve blocks could provide sufficient analgesic benefits to facilitate lengthy transport of lower-extremity polytrauma injured soldiers from the battlefield to definitive care. The study was awarded funding, with the original dates being October 2015 to September 2019. Unfortunately, we were confronted by several enrollment and timeline challenges, and, despite receiving a study extension, were unable to reach our enrollment milestone. The target sample size for the approved randomized clinical trial was a total of 200 patients, with 100 undergoing total hip arthroplasty and 100 total knee arthroplasty. This target sample size was not obtained because of regulatory challenges, limited collaboration from orthopedic surgeons, and changes in policies that affected surgical eligibility. The regulatory challenges included that the institutional review boards of three institutions had oversight of the protocol: VA Pittsburgh, University of Pittsburgh, and Department of Defense. The most labor-intensive institutional review board requirement involved Food and Drug Administration Investigational New Drug application and approval. To our knowledge, no Food and Drug Administration Investigational New Drug approval was required on any previous or subsequent study of two or more of the BCD adjuvants combined with local anesthetics. Before the study could begin, the Food and Drug Administration required a clinical safety chart review of our nerve block cases from 2011 to 2014 [3], comprising more than 1,800 patients receiving BCD. These regulatory requirements delayed the study start by 6–8 months; recruitment was originally scheduled for March 2016, but because of the regulatory requirements, the first patient was not actually enrolled until November 2016. The study was also negatively impacted by limited collaboration from orthopedic surgeons. The recruitment process for this study involved more than 2 weeks of interactions with potential participants to enable sufficient time to mail study information, schedule appointments, and bring participants back to the hospital for screening and baseline procedures ahead of the surgical date. However, the surgeons did not have sufficient time to discuss the study with their patients. Other surgeons scheduled the surgery to occur in less than 2 weeks from initial clinic presentation, for pragmatic reasons. Moreover, the orthopedic practice instituted a policy to not operate electively on patients with positive urine cotinine tests. This was a major hurdle because of the prevalence of veterans’ tobacco use. Lastly, patient enrollment required substantial effort and resources: Only 17% of screened patients (98 of 571) were eligible to participate because of these timing complexities and other exclusions, reinforcing that regardless of the trial sponsor, recruitment of patients who meet the requirements of clinical trial protocols can be difficult [4]. In mid-2018, the VA Pittsburgh adopted the evolving nationwide guideline of same-day physical therapy after elective joint arthroplasty, which was first proposed by the American Academy of Orthopaedic Surgeons in December 2015. Processes shifted from initial patient mobilization occurring on the first postoperative day (POD#1) to instead occurring immediately after surgery (POD#0). As a result, the Data Safety Monitoring Board suspended study enrollment until we diluted the nerve block local anesthetic concentrations to enable patients’ safe engagement in same-day physical therapy and gained all necessary re-approvals. At the time of study suspension, 98 veterans (48% of the target sample) had been enrolled in the study. The Data Safety Monitoring Board recommended that we obtain all regulatory approvals and continue study enrollment under the revised study protocol to obtain valuable preliminary data for this new standard of care of starting physical therapy on POD#0. Because of the differences in drug concentrations, the Data Safety Monitoring Board also suggested we not combine data from the two study cohorts but rather provide descriptive statistics of the second cohort and visually compare them with the cohort enrolled before study suspension. We used our first cohort data of 98 participants for our first Pain Medicine article from this study population, addressing analgesic duration from the described nerve blocks [5]. The regulatory approvals kept the study suspended for nearly 7 months, until late June 2019. Because the original funding period went through September 2019, we proposed a no-cost extension until March 2020 to enroll the new cohort. However, recruitment efforts did not fare much better, resulting in 40 additional participants in the second cohort. As before, challenges included the surgical team scheduling surgery in immediately available time slots created by the negative urine cotinine requirement, very few days after the veteran initially presented to the clinic. On a positive note, unrecruited potential study subjects were routed into same-day mobilization anesthesia clinical pathways, which provided other opportunities to further evaluate BCD nerve blocks, resulting in related publications in Pain Medicine [6–9]; these reports inform future research directions in ways we could not have with the prospective randomized trial. With the change in nerve block methodology away from the originally planned, potent BPV-BCD (e.g., with 0.25% BPV concentration in the L2–L4 distribution), the next inquiry entails MMPNA interventions (i.e., drugs other than local anesthetics) as proximal nerve/plexus motor-sparing blocks, to better position patients for POD#0 mobilization. Since 2013, a plethora of ultrasound-guided motor-sparing distal-to-plexus block techniques have been described, which are beyond the scope of this editorial and accompanying article [1]. In the meantime, one could consider combined spinal anesthesia with both motor-sparing lumbar plexus psoas compartment and L4-S3 distribution analgesic blocks. Specifically, lumbar plexus analgesia would be with either midazolam or ultra-low-concentration BPV, combined with BCD [7, 9], while L4–S3 analgesia entails midazolam-BCD blocks of the sciatic nerve for total knee arthroplasty, and of the parasacral plexus for total hip arthroplasty. These motor-sparing blocks allow for both analgesia and same-day mobilization, ultimately involving two complex nerve stimulator techniques and three overall injections (with the third injection being the spinal anesthetic). Another option worth considering for total knee arthroplasty is lumbar plexus psoas compartment and sciatic blocks (a two-needle, complex nerve-stimulator injection technique), entailing a combination of shorter-acting local anesthetics (e.g., mepivacaine with or without dilute BPV) and BCD to promote a surgical block without general or spinal anesthesia and rapid (∼4–6 hours from time of block) motor resolution to allow same-day physical therapy [8]. These three described L2–S3 blocks (L2–L4 lumbar plexus psoas compartment, L4–S3 parasacral plexus, and L4–S3 Labat-approach gluteal sciatic) do not require ultrasound guidance but do require training and experience with landmark-based peripheral nerve stimulation. The simplest, most motor-sparing answer with effective analgesia may lie in the single-injection spinal anesthetic, with a mixture of surgery-duration local anesthetic and longer-duration preservative-free morphine, for which a meta-analysis has recently been published [10]. Our clinical observations [9] of intrathecal morphine with regard to successful same-day ambulation with a low-burden POD#0 pain/opioid outcome have proved to be quite promising. The new clinical research question now lies in how many antiemetic categories should be used to offset the potently emetogenic intrathecal morphine [6, 10], as opposed to how many MMPNA preservative-free drugs to inject at the level of the plexus or proximal nerve. Appropriate concentrations of BPV (e.g., 0.25%) combined with the described [1, 2, 5] doses of BCD will provide 24–36 hours of analgesic duration [2, 5], translating to less pain [1]. With regard to anesthesia for elective joint arthroplasty, the described nerve stimulator blocks have likely become an intergenerational lost art since 2013, when we first submitted this grant application. Ultimately, with appropriate multiple-agent antiemetic prophylaxis [6], intrathecal morphine for joint arthroplasty may prove equal or superior to peripheral/plexus blocks with ultrasound and/or nerve stimulation, with local anesthetic and/or MMPNA mixtures, with or without local anesthetic infiltration by our orthopedic colleagues. Funding sources: This commentary is based on the study underwritten by a grant supported by the Office of the Assistant Secretary of Defense for Health Affairs through the FY14 DoD USAMRMC Broad Agency Announcement under Award No. W81XWH-15–1-0294. Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the Department of Defense or the Veterans Health Administration. Conflicts of interest: The authors report no conflicts of interest.