As in other types of major musculoskeletal surgery, pain control following lumbar decompression and arthrodesis can be challenging. Parenteral narcotics, usually with a patient-controlled pump, remain the mainstay. While effective for most patients, there are substantial problems with this modality, including respiratory depression, nausea, emesis, and pruritus. Some patients, particularly those who are not opioid naïve, have perioperative pain that is refractory to narcotic analgesia. There is long-standing interest in techniques to reduce our dependence on opioids. The report by Hyun Kang et al. is one of the latest in a long line of studies examining regional blocks to reduce the need for parenteral narcotics.
This study has many strengths. It is randomized, double-blinded, prospective, and controlled. All patients had the same surgical procedures at the same institution, minimizing variation in the amount of surgical trauma. The study and control groups were statistically equivalent. The investigators demonstrated that a single dose of epidural ropivacaine reduced pain scores and fentanyl consumption two to twelve hours following surgery. The number of button pushes to deliver fentanyl was decreased at all but one time point up to forty-eight hours postoperatively. However, there was no difference between the two groups in terms of postoperative nausea and emesis, the length of hospitalization, or patient satisfaction. It should be noted that the study design was powered to discern the 2-point difference in pain at four hours. The study may be underpowered concerning these secondary but interesting parameters. Six patients had to be replaced because of nausea, emesis, or delirium, demonstrating that neither experimental nor standard treatment was successful in everyone.
Other regional techniques yield similar outcomes. France et al. used long-acting intrathecal morphine in another prospective randomized double-blind study of sixty-eight patients who were managed with posterolateral lumbar arthrodesis1. The forty-two patients who were managed with intrathecal morphine had less pain than the twenty-six controls for the first twenty-four hours, although the difference was significant only at time zero. Patient-controlled analgesia use was significantly less in the experimental group throughout the first twenty-four hours. Interestingly, there was rebound, with more pain and patient-controlled analgesia use in the treatment group after twenty-four hours. Three patients who were managed with intrathecal morphine had mild respiratory depression, and there was more pruritus in that group. The limitations of this method include the inability to repeat the injection, the rebound phenomenon, and potential spinal fluid leakage.
Prasartritha et al. compared epidural morphine (administered with or without bupivacaine) with intravenous morphine in a study of 120 randomized patients undergoing lumbar decompression, arthrodesis, or both2. Visual analog scale pain scores were lower in the epidural groups as compared with the parenteral drug group up to forty-eight hours. Patients managed with bupivacaine had less pain than those managed with epidural morphine alone at sixteen and twenty-four hours. There was no respiratory depression, but pruritus was common in the epidural groups. This method allows additional dosing but has the potential disadvantages of catheter misplacement, leakage, infection, epidural bleeding, and neural complications from the catheter.
Diaz et al. performed a randomized double-blinded controlled trial of 201 patients who were managed with lumbar discectomy or decompression for the treatment of spinal stenosis3. They prepared four different epidural pastes: (1) a combination paste (morphine and methylprednisolone), (2) a steroid paste (methylprednisolone alone), a morphine paste (morphine alone), or placebo (microfibrillar collagen alone). Significant reduction in pain ratings in the first three days after surgery occurred in association with the combination and steroid paste but not in association with morphine or collagen alone. The same held true for narcotic analgesic consumption. Although the dosage of the active ingredients varied between the two types of decompression, the authors speculated that the morphine dosage may have been too low. Secondary clinical outcome measures were similar in all groups up to one year. The technique has the potential to retain the medication at the surgical site but does not allow repeat dosing. It avoids the disadvantages of an epidural catheter without violating the dura.
There are myriad similar studies in the literature. Clearly, regional analgesia can successfully reduce parenteral narcotic requirements in the early postoperative period. Each technique has advantages, disadvantages, and potential side effects. The study by Kang et al. adds to the options. The existence of many viable interventions raises an obvious question: why are none of these techniques widely used? Kang’s method is simple and could be implemented if the anesthesiologist were skilled in epidural drug delivery and could do so efficiently. The authors did not indicate how much total additional anesthetic time was required. There was a twenty-minute period between injection and incision. The total time of injection plus the twenty minutes adds additional expense of operating room time. Does the reduction in pain and parenteral opioid consumption justify this? In this pilot study, there was no reduction in hospitalization, which could offset this additional expense. The authors may wish to pursue additional studies examining these questions.
It is intriguing that Kang and colleagues demonstrated a significant blunting of C-reactive protein elevation in the experimental group, which peaked at three days in both groups. They postulated attenuation of the inflammatory response but admitted the need for additional study. The Diaz study supports this hypothesis in that pain reduction was most significant in the groups receiving paste containing methylprednisolone. Further investigation of the inflammatory response is definitely warranted.
At this point, the interested spine surgeon can partner with anesthesiologists to reduce parenteral opioid requirements in a number of ways but still lacks data that any of them are cost-effective or yield better clinical outcomes. The opening sentence in the introduction of Kang’s study caught my attention. The authors implied that postoperative pain management could affect persistent chronic postoperative pain. Is this true? Kang et al. provide no long-term follow-up data. Diaz et al. found no differences in validated clinical outcome measures at one year, but these were secondary outcomes. I would strongly encourage Kang and associates to perform a larger study of their technique with at least one and preferably two-year follow-up for the evaluation of pain and patient outcome measures. They should also assess the difference in cost between the two methods. If such data show supplementary regional analgesia to be superior in terms of outcome at comparable cost, surgeons are more likely to adopt these techniques.
I congratulate Kang and colleagues on an excellent preliminary study. I urge them to continue and expand their work.