Spinal Anesthetic for Hip and Knee Replacement Surgery - Intrathecal Morphine or Hydromorphone

Overview

In the current healthcare system, a short length of stay is paramount, for medical, functional, and financial reasons. Most patients undergoing hip and knee replacement at Hôpital Montfort institution receive spinal anesthetic, with intrathecal morphine for enhanced analgesia. However, morphine can cause urinary retention, nausea or vomiting resulting in increased length of stay (LOS). Based mainly on local experience, The investigators believe hydromorphone may be a good alternative (1). According to the literature review, no adult study compares morphine and hydromorphone. QUESTIONS Does intrathecal hydromorphone lead to equal or better pain control/post-operative function, with fewer side effects than morphine? OBJECTIVES. Primary objective: to demonstrate that hydromorphone is the choice narcotic for spinal anesthesia for arthroplasty patients. Secondary objectives: * increased patient satisfaction * decreased hospital length of stay METHODOLOGY The investigators propose a prospective randomized double-blind trial with three groups of 155 patients each undergoing total hip arthroplasty (THA), total knee arthroplasty (TKA), or unicompartmental knee arthroplasty (UKA), striving for similar proportions of each procedure in all three groups. * Group 1: spinal anesthetic with standard painkillers, without intrathecal narcotics (ie "placebo" group) * Group 2: spinal anesthetic with intrathecal morphine * Group 3: spinal anesthetic with intrathecal hydromorphone The primary outcome measure will be urinary retention. Secondary outcome measures will be nausea/vomiting, pain control, post-operative function and length of stay. RESULTS The investigators expect to show that intrathecal hydromorphone will provide the best combination of post-operative patient pain control and function, with diminished negative side effects. They hope to extrapolate this to a potential out-patient arthroplasty program.

A significant increase in the number of patients receiving hip or knee replacements can be expected in the coming years, for two main reasons: an increase in the number of elderly patients in the society and the fact that joint replacements are offered to younger patients because of an improvement in materials, designs, and survivorship of prostheses in general. As a result, healthcare institutions are facing pressure to provide more services, while decreasing overall spending. Hôpital Montfort is an important component of the healthcare structure in Ottawa and Eastern Ontario, providing hip or knee replacements to roughly 1200 patients every year. These include total hip and knee replacements, partial medial and lateral knee replacements, and patellafemoral replacements. Some partial knee replacement surgeries are bilateral, meaning both knees are operated at the same sitting.An increase in the demand for joint replacements is expected. This will follow the national trend of the last 5 years (2012-2017): 17.8% increase in hip replacements and 15.5% increase in knee replacements. This in turn will continue to place an unbearable financial burden on the healthcare system, as implanted components are very expensive. Furthermore, patients undergoing joint replacement also often require the use of ancillary services such as physiotherapy, occupational therapy, and inpatient rehabilitation. However, the healthcare system is under tremendous strain, as budgets can no longer allow previous levels of spending. It is this current economic context that is forcing healthcare professionals to find ways to decrease costs as much as possible, while maintaining productivity. Over 95% of joint replacements at Hôpital Montfort are performed under spinal anesthetic. Intrathecal morphine is added to standard anesthetic medication for better immediate post-operative pain control, which in turn facilitates rapid patient mobilization. However, most orthopedic surgeons would prefer to avoid morphine, because of a recognized higher incidence of post-operative side-effects, mainly urinary retention, nausea, and vomiting. It is very difficult to send patients home when they cannot urinate, tolerate oral intake, or mobilize safely. Patients who experience one of these post-operative complications stay in hospital. Discharging these patients home can be unsafe, but also leads to patient dissatisfaction, and the perception of uncaring and unsympathetic care. Many patients do not understand the financial implications of a prolonged stay in hospital, and expect that they can stay as long as they want. A medication that can meet all our patients' needs: effective analgesia, early mobilization, and minimal negative side effects, is still to be found, so patients can go home sooner, The anesthesia literature seems to indicate that hydromorphone might be the right choice, but no study compared intrathecal morphine and hydromorphone usage in the adult population. QUESTIONS * Will the use of intrathecal hydromorphone (compared to morphine and placebo) for patients undergoing joint replacement: * diminish the incidence of urinary retention requiring catheterization? * diminish the incidence of nausea and vomiting? * be associated with equivalent or better pain control? * be associated with equivalent or better post-operative function and mobilization? * allow earlier discharge, with shorter LOS? * cause unpredicted negative outcomes, such as higher rate of post-operative emergency department visit or readmission before first post-operative visit? * Will patients in thecontrol group do as well as patients in the intrathecal narcotic groups, with respect to pain control and mobilization, as well as negative secondary side effects? OBJECTIVES * Main objective: compare intrathecal hydromorphone to morphine and control, looking specifically for differences in post-operative negative effects (urinary retention, nausea and vomiting, pain levels, mobilization and function) * Secondary objectives: * Patient specific: improve patient experience by offering improved spinal anesthetic that will provide the best possible patient experience by facilitating recovery and decreasing post-operative complications related to anesthetic medications. * Hospital-specific: identify which spinal anesthetic combination of medications will allow earliest discharge from hospital, thereby saving money for the hospital * System-specific: identify which spinal anesthetic combination of medications will allow us to extrapolate this to a potential out-patient total joint replacement program. HYPOTHESIS * Spinal anesthetic with intrathecal narcotic has been shown to increase post-operative urinary retention, with rates ranging from 40% (14) to 60%, and even as high as 70% in one review article. Some centres are therefore avoiding intrathecal narcotics altogether, and finding non-inferior results by adding adductor canal blocks and local infiltration to the spinal. Based on a preliminary review of the literature, the investigators expect to show that intrathecal hydromorphone is associated with less urinary retention, nausea and vomiting, than morphine, without jeopardizing pain control and patient mobilization. This will allow to exploit the positive analgesic benefits of intrathecal narcotics, without risking the possible side effects associated with the large volumes of local anesthetic required for blocks and peri-articular infiltration. * Therefore a shorter lenght of stay is expected. * Bboth forms of intrathecal narcotic to provide better pain control than spinal without narcotic. * Spinal without narcotic leads to lower incidence of urinary retention, nausea and vomiting. METHODOLOGY Design and procedures: Subject to hospital ethics review, the investigators will conduct a prospective randomized double-blind study of patients undergoing total hip or knee replacement, as well as partial knee replacements (medial, lateral, patellafemoral, or bilateral). All groups will be matched for age, sex, American Society of Anesthesiologists class, and type of surgery. An attempt will be made to achieve a similar proportion of THA, TKA, and partial knee replacements in each group. Patients will be assigned to one of the following groups: 1. Patients receiving spinal anesthetic with standard painkillers without intrathecal narcotics 2. Patients receiving spinal anesthetic with 100 micrograms of intrathecal morphine 3. Patients receiving spinal anesthetic with 20 micrograms of intrathecal hydromorphone (which has a potency of 5:1 relative to morphine) All the prepared syringes will be 0.2 ml in volume. Medication syringes will be prepared ahead of time by Pharmacy, and distributed according to the randomization process (see below). The surgeon will follow his or her regular surgical technique, while the anesthetist will be responsible for all aspects of anesthetic care. The investigators will follow our pre-established hip and knee protocols, which cover pre-medication (prophylaxis for pain, nausea, vomiting), post-operative management, physiotherapy and occupational therapy, etc. Recruitment and randomization: The sample size for this study was determined based on analyses for the primary outcome of urinary retention. As mentioned above, previous studies have reported variable incidence rates of post-operative urinary retention and catheterization for arthroplasty, anywhere from 0 to 70%. Although statistics for 2017-2018 at Hôpital Montfort revealed a relatively low incidence rate (less than 10%), the lack of consistent diagnostic and reporting criteria render this value unreliable (and likely underreported) for sample size computations. The investigators therefore assumed a more realistic incidence of approximately 30% consistent with previous studies, taking into account factors such as sex, surgery type, anesthesia, etc. Using GPower, the investigators calculated that 155 patients would be required in each study arm to show a 50% percent reduction in the rate of urinary retention between treatment groups (two-tailed). This number of subjects (465 in total) would give 80% power at the 0.02 level (.05 divided by 3 comparisons). The Hôpital Montfort Arthroplasty Clinic physiotherapist assessors, who do the preliminary assessments and education for all of patients, will identify eligible patients. Consent to participate in this study will not be obtained until patients have met first with their orthopedic surgeon (to go over the surgical procedure and post-operative expectations), and then seen an anaesthetist in the pre-admission clinic to go over the anaesthetic protocol, and the three anesthetic options associated with this study. The consent form will be signed in the pre-admission clinic, and witnessed by a nurse or the anaesthetist consultant, once the patient expresses understanding and acceptance of the principles of this study. Patients will be randomized into the three groups by computer-generated random assignment using SPSS software. Allocations will be generated using permuted blocks of randomly varying lengths of 3 or 6 to maintain close balance of the numbers of patients in each treatment group at any time during the trial and to ensure allocation concealment. Primary and secondary analyses. All data will be entered and analyzed in SPSS version 22. Data will be cleaned and summary statistics will be computed, notably means, medians and standard deviation for continuous variables and frequencies (%) for categorical variables. The primary outcome of urinary retention will be analyzed via chi square test with p \< .02 (adjusted for multiple comparisons). A backward stepwise logistic regression model yielding odds ratio (OR) and 95% confidence interval (CI) will be performed to analyze the treatment effect when adjusting for possible covariates (e.g., age, sex, type of surgery, post-operative care, etc.). The goodness of fit of the regression model will be evaluated by the Hosmer-Lemeshow test. Secondary binary outcomes (i.e., incidence of nausea and vomiting) will be analyzed in the same fashion. The strategy for analyzing count variables (e.g., episodes of vomiting, analgesics) will depend on the distribution of the data, though the chi-square test will be the preferred initial approach to examine the unique effect of group (two-tailed, alpha adjusted for multiple comparisons). A negative binomial regression model will allow for examining the contributing influence of possible covariates listed above. The effect of treatment group on continuous outcome variables (e.g., length of stay, patient questionnaire scores) will initially be analyzed via one-way Analysis of Variance (ANOVA) with post-hoc pairwise comparisons (with Bonferroni correction). Non-parametric tests (i.e., Mann-Whitney, Kruskal-Wallis) will be considered for non-normally distributed data. Despite some contention regarding the analyses of visual analogue scales (i.e., pain measurement), non-parametric approaches seem to be favoured and therefore a Kruskal-Wallis test will be applied (with post-hoc pairwise comparisons). Next, multiple step-wise linear regression models will be considered to examine the effect of possible baseline covariates.