Patients with a history of migraines are at increased risk for hospital readmission due to pain within 30 days. A database analysis concluded that migraineurs are more frequently readmitted for postoperative pain, headache and abdominal pain than their counterparts without migraines.
“We are all familiar with the fact that unplanned hospital readmissions are of great interest these days,” said Meghan B. Zhang, BS, who is a medical student at the University of Maryland School of Medicine, in Baltimore, but who was part of the research team while at Massachusetts General Hospital, in Boston.
“They can be used as metrics of hospital performance, as well as the quality and cost of patient care. So it’s important that we work towards identifying what types of readmissions can be prevented and come up with solutions for reducing their occurrence,” she said.
To that end, the researchers studied the records of 122,494 adult surgical cases presenting to the tertiary care center and two community hospitals between 2007 and 2014. Patients with a history of migraines—with or without aura—were defined using International Classification of Diseases, Ninth Revision diagnosis codes. The primary outcome of the analysis was hospital readmission for pain within 30 days of surgery.
Presence of Aura Increases Readmission Rate
As Ms. Zhang reported at the 2017 annual meeting of the American Society of Anesthesiologists (abstract A3008), 10,009 (8.17%) patients were diagnosed with migraine. Of these, 1,257 (1.03%) had migraine with aura. “Our migraineurs tended to be younger and were more likely to be women,” Ms. Zhang said. In total, 650 patients (0.53%) were readmitted within 30 days of surgery with an admitting diagnosis specifying pain.
The investigators found that patients with a history of migraine were at higher risk for readmission due to pain (adjusted odds ratio [aOR], 1.74; 95% CI, 1.39-2.18; P<0.001). Furthermore, migraine patients with aura were at higher risk for pain-related readmissions (aOR, 2.82; 95% CI, 1.80-4.43; P<0.001) than their counterparts without aura (aOR, 1.58; 95% CI, 1.80-4.43; P<0.001).
The predicted event rates for pain-related readmissions among non-migraineurs and migraineurs with and without aura were 432 (0.38%), 66.6 (0.67%) and 108 (1.08%), respectively. Among readmissions due to pain, migraineurs were more likely to be admitted for postoperative pain (aOR, 3.33; 95% CI, 1.51-7.34; P=0.003), headache (aOR, 3.12; 95% CI, 1.76-5.54; P<0.001) and abdominal pain (aOR, 1.93; 95% CI, 1.43-2.60; P<0.001).
Ms. Zhang and her co-investigators also performed exploratory analyses targeting the effect of migraine medications on pain-related readmissions in the cohort population. They found a nonsignificant trend indicating that migraine patients who did not have a prescription for migraine-abortive medications may be more vulnerable to readmission for migraine-associated pain after surgery, compared with any other migraine patient in the cohort.
“So, in conclusion,” she said, “we identified migraineurs—especially those with aura and those with chronic migraines—as vulnerable perioperative patient populations in terms of pain management,” Ms. Zhang said. “Efficient pain prevention and treatment in surgical migraine patients can lower the burden of postoperative readmission, and the provision of migraine-abortive treatment for episodic migraine patients is a potential intervention to prevent postsurgical readmission.”
Despite the strength of these findings, not all of Ms. Zhang’s audience members were convinced. “How does the [Massachusetts General Hospital] database flag the presence of a preoperative migraine?” questioned Nathan L. Pace, MD, a professor of anesthesiology at the University of Utah, in Salt Lake City. “I ask because you said there’s a 15% incidence of migraine in the general U.S. population, but your database shows a migraine incidence of only about 6.5%. That’s quite a discrepancy.”
“We believe it was a conservative error,” replied co-investigator Matthias Eikermann, MD, a professor of anesthesia at Harvard Medical School, in Boston. “For if you can identify such an association even though some patients were misclassified as normal even though they had migraine, you can assume that in all likelihood we’re presenting a conservative error.”