Cutting Edge / Controversies
The initial warning statement in Cope's Early Diagnosis of the Acute Abdomen against the use of pain medication in patients with abdominal pain was not evidence-based, yet became dogma that has been hard to overcome. 125 At the time, 1921, it was not unusual to use morphine in doses of 30 mg, thus, it was more likely to render the patient difficult to arouse, let alone examine. 126
There are now multiple studies in the literature that contradict this dogma. Zoltie and Cust performed a prospective, double-blinded study in 1986 comparing buprenorphine sublingual versus placebo. The physical exam did change; however, in 18 of 50 patients studied with a change in exam, the change involved a localization of pain which facilitated the diagnosis. They concluded that buprenorphine can be safely given to patients with an acute abdomen. 127
In 1992, Attard et al conducted a prospective, randomized, placebo-controlled study evaluating intramuscular papaveretum (dose equivalent to morphine 12.5 mg) versus placebo. VAS and abdominal exam were done prior to medication and one hour later. The study assessed the examiner's confidence in diagnosis and management decisions before and after treatment. Median pain and tenderness scores fell significantly after papaveretum versus placebo, with no change in localization of the tender area. Interestingly, there were six unnecessary operations in the placebo group and none in the study group.
There were two patients in the study group who were discharged and subsequently diagnosed with appendicitis. The surgical registrar's confidence in diagnosing and making management decisions were not altered by papaveretum. They concluded that early pain relief with papaveretum in patients with severe acute abdominal pain did not have any adverse effect on diagnosis. 128
In 1996, Pace and Burke evaluated IV morphine compared to placebo in patients with acute abdominal pain. They looked at changes in pain scores before and after drug administration and at the accuracy of the initial diagnosis compared with the final diagnosis. The morphine group reported significantly less pain. Treatment with morphine did not result in resolution of peritoneal signs. The primary and final diagnosis agreed in 80% of patients in the morphine group and in 61% of patients in the placebo group. Three patients in each group were misdiagnosed. The authors concluded that administration of morphine to adult patients with acute, atraumatic abdominal pain resulted in pain relief without altering the ability of emergency physicians to perform accurate evaluations. 129
In 1997, LoVecchio et al studied morphine (5 mg and 10 mg) compared to placebo in patients with abdominal pain. The study evaluated changes in the physical exam, including changes in tenderness from two or more quadrants to one, or loss of rebound tenderness, and vice versa. Their results showed a significant reduction in pain after both low and high doses of morphine. Nine of 19 patients changed their physical exam after high dose morphine, seven of 13 patients with low dose morphine, and only one of 16 patient in the placebo group. There was no morbidity or delay in patient care. The authors concluded that early administration of opiate analgesics is safe and effective in patients with acute abdominal pain in the ED. The authors theorized that the change in exam was due to muscular relaxation with decreased guarding with better localization. 126
Finally, in 2003, Thomas et al performed a prospective, randomized trial evaluating adult patients with undifferentiated abdominal pain, comparing placebo with morphine. They concluded that there was no "untoward effect" with early administration of morphine, and "No evidence supporting the contention that [morphine] administration was deleterious in any way." 130
Despite the fact that these five randomized, double- blinded studies support the use of analgesics in non-specified abdominal pain, clinical practice has lagged. In 1999, Tanabe found that only 35% of patients with abdominal pain received an analgesic.10 A survey of surgeons by Graber et al in 1999 found that 53% of general surgeons "Believe that pain medications preclude a patient from signing a valid informed consent" and 67% believe that "Narcotics can hinder diagnostic accuracy." 131 Wolfe et al surveyed emergency physicians in 1997 and found that, while 85% of physicians felt that pain medication did not change "important physical findings," 76% of physicians chose not to administer narcotics until the patient was seen by a surgeon. Factors associated with whether or not medication was given included: The degree of pain that the patient was experiencing (69%), the certainty of diagnosis (52%), and the length of time until surgical evaluation (58%). 132
In 2003, Kim et al surveyed pediatric emergency (PEM) physicians and pediatric surgeons and found that, in physicians with greater than 10 years experience, pediatric surgeons were less likely than PEM - physicians to give narcotics (61% vs. 38%), with PEM physicians citing disapproval by pediatric surgeons as the main barrier for prescribing analgesics. 133 This did not hold true for physicians with less than 10 years experience. This likely speaks to the fact that physicians who have trained in the last 10 years trained during the time when the dogma of with- holding pain medication in the evaluation of abdominal pain has been questioned and refuted. The availability and use of CT scan for the diagnosis of abdominal pain likely also contributes.
An audit by Tait et al in 1999 found that 43% of patients with abdominal pain were admitted to the hospital without receiving an analgesic in the ED and had an average wait time of 5.7 hours for pain medication. 134 Similarly, a prospective observational study by Shabbir et al in 2004 of patients presenting to the ED with abdominal pain demonstrated that 67% of patients received pain medicine within one hour, but 22% waited from two to 14 hours. Those patients with less severe pain waited longer and female patients also had a longer wait than men, mean 129 minutes versus 69 minutes respectively. 135
These last two studies demonstrate that, while the literature supports the use of narcotic medication in the evaluation of abdominal pain, clinical practice has not yet changed. There are still a significant number of patients admitted to the hospital who do not receive pain medication, and patients who wait long periods of time prior to pain medication. Developing collaborative protocols for the management of acute, non-traumatic abdominal pain would seem prudent. On a final note, capacity to give informed consent for potential surgical interventions occasionally emerges as a weak argument against initiating pain treatment prior to an evaluation by a surgeon.
However, it can be argued that pain itself clouds judgment, that patients in severe pain can think of nothing but their pain and will agree to anything for the relief of pain. 136 Hence, it would be coercive to withhold pain medication until informed consent has been signed. Physicians should document, when appropriate, the necessity of giving pain medications before consent and the patient's cognitive state when deciding, or else document that the delay in paincontrol was necessary and tolerable in order to obtain proper consent.
Intramuscular Ketorolac Versus Oral Ibuprofen NSAIDs inhibit prostaglandin synthesis and thus are especially useful in pain syndromes where pain is prostaglandin-mediated, such as dysmenorrhea, biliary colic, renal colic, arthritis, postoperative pain, pharyngitis, and soft tissue trauma. 137 Some physicians prefer to use IM toradol rather than oral ibuprofen. However, IM dosing does not have a faster onset of action than oral dosing, and has several potential complications. Wright et al published a retrospective analysis of data collected during a prospective study on pain management in which ketorolac, 60 mg IM, was compared to ibuprofen, 800 mg PO. A VAS tool was used to assess pain pre- and post-treatment. There was no statistical difference in pain relief by change in VAS score or by descriptive assessment. In fact, approximately 40% of patients in both groups described pain relief as none or little. 138
In a randomized, prospective double-blinded study, Turturro et al compared IM ketorolac (60 mg) and oral ibuprofen (800 mg) in acute musculoskeletal pain using a VAS instrument.137 The level of pain was evaluated at baseline, 15, 30, 45, 60, 75, 90, and 120 minutes after dosing. Mean pain scores did not differ etween the groups and they concluded that IM ketorolac and PO ibuprofen provides similar analgesia. A third study looking at IM ketorolac and oral ibuprofen was done by Neighbor and Puntillo. 139 It was a prospective, randomized, double-blinded study of patients with moderate to severe pain using a NRS tool to assess pain at baseline, 15, 30, 45, 60, 90, and 120 minutes after treatment. Supplemental analgesics were allowed. No difference was noted between the two groups. An additional study by Shrestha et al compared IM ketorolac and oral indomethacin in the treatment of acute gouty arthritis. This prospective, randomized, double-blinded, controlled study found no difference in pain relief. 140 There has been concern that the above studies failed to control for the presumed placebo-effect that an IM injection has on a patient's perceived pain relief; however, this would only further favor the use of oral dosing over IM. Regardless, one study did attempt to control for this potential placebo effect by using saline injections in the group receiving oral ibuprofen and starch tablets in the group receiving IM ketorolac; no difference in pain relief between oral and IM dosing was found. 78
It can be concluded from these studies that there is no significant pain relief benefit of using IM ketorolac instead of oral ibuprofen. In that IM dosing has theoretical risks and is more expensive, the use of oral ibuprofen is clearly the best choice. However, when the parenteral route is required (i.e., the patient is vomiting), there is a role for IM toradol.
Cox-2 inhibitors were developed and introduced in an attempt to find an anti-inflammatory drug that would not carry the anti-thrombotic properties of the nonspecific cyclooxygenase inhibitors. From the time of their introduction, refocoxib (Vioxx™ from Merck) and celecoxib (Celebrex™ from Pfizer) grew to incredible popularity. In 2003, Cox-2 inhibitors were the seventh largest selling class of drugs, with over five billion dollars in annual sales. 141 Eric Topol published a now famous challenge to the safety of Cox-2 inhibitors in JAMA in 2001 asserting that Cox-2 inhibitors did not share the same anti-thrombotic properties of non-selective Cox inhibitors, and that there was an increased risk of cardiovascular events. 142 Upon review of the studies submitted to the FDA for approval of the Cox- 2 inhibitors, he discovered that patients treated with refocoxib in the Vioxx™ Gastrointestinal Outcomes Research Study (VIGOR; 8076 patients) had an increased risk of adverse cardiovascular events, including myocardial infarction and ischemic stroke, than patients treated with naprosyn. The Celecoxib™ Long-term Arthritis Safety Study (CLASS; 8059 patients) did not find an increased risk of adverse cardiovascular events when celecoxib was compared to ibuprofen and diclofenac (Voltaren™).142 The difference in the VIGOR study and the CLASS study accounts for the difference in the legal troubles that have been faced by Merck and Pfizer.
While this has made headlines many times over, what goes unaddressed is the problem facing patients who had found the COX-2 inhibitors to be a great source of pain relief. It remains unclear what are the best alternatives for these patients; several authors have recommended the use of traditional NSAIDs with gastro-intestinal protection either with a proton-pump inhibitor or a histamine-2 receptor antagonist (H2RA). 143, 144 Elliott has suggested that treatment with a H2 receptor antagonist is a cost effective way to communicate the benefits of NSAID pain relief to patients while minimizing the risks of gastrointestinal complications. 143 Other authors have advocated the continued use of COX-2 inhibitors in select patients after careful screening. Given the current medico-legal climate, this could leave the prescribing physician in a precarious legal position and careful documentation of counseling is recommended.