1137

Low Back Pain Caused by a Duodenal Ulcer David J. Weiss, MD, Theodore ConlifSe, MD, Narayan Tata, BS

ABSTRACT. Weiss DJ, Conliffe T, Tata N. Low back pain caused by a duodenal ulcer. Arch Phys Med Rehabil 1998;79: 1137-9.

The common diagnoses in low back pain are lumbar strain, lumbosacral radiculopathy, osteoarthritis, degenerative disc disease, spinal stenosis, and sacroiliac joint dysfunction, Un- usual causes of low back pain that have been previously identified include abdominal aortic aneurysms, pelvic neo- plasms, and retroperitoneal hemorrhages. This report describes a case of back pain that was apparently caused by a duodenal ulcer. A 54-year-old man with no significant medical history presented with a complaint of mid to low back pain (TlO-L2), which was diagnosed as joint dysfunction. A comprehensive treatment program was prescribed and the patient was in- structed to return to clinic in 4 weeks. Three weeks later, he experienced a syncopal episode followed by coffee ground emesis. He immediately sought medical attention at an emer- gency room, where he was admitted to the hospital with a diagnosis of upper gastrointestinal bleed. Esophagogastroduo- denoscopy showed a large duodenal ulcer, and the patient underwent vagotomy and pyloroplasty. He returned to his physiatrist’s office 3 weeks after hospital discharge with minimal back pain. The cause of the back pain proved to be referred visceral pain from his duodenal ulcer. This case is presented to reemphasize the need to include the uncommon phenomena in the differential diagnosis of low back pain.

0 1998 by the American Congress of Rehabilitation Medi- cine and the American Academy of Physical Medicine and Rehabilitation

M OST AMERICANS will experience low back pain at some point in their 1ife.l It is the second most common

reason why patients see their physicians (the first is upper respiratory infection). The diagnoses that are most often used when patients present with low back pain include lumbar sprain/strain, lumbosacral radiculopathy, osteoarthritis, degen- erative disc disease, spinal stenosis, facet joint syndrome, and sacroiliac dysfunction. Recently, more of the uncommon causes of low back pain have been reported. A few of these uncommon causes include abdominal aortic aneurysms2 hematometra,3 and retroperitoneal hemorrhages.4 We present a case of low back pain that was apparently caused by a duodenal ulcer.

CASE REPORT A 54-year-old man with an unremarkable medical history

presented with a 3-month history of mid-to-low back pain. He

From the Schwab Rehabilitation Hospital & Care Network (Drs. Weiss, Conliffe); University of Chicago Hospitals, Division of Orthopaedic Surgery & Rehabilitation Medicine (Dr. Weiss); and Chicago Medical School (Mr. Tata), Chicago, IL.

Submitted for publication January 12, 1998. Accepted in revised form April 13, 1998.

No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.

Reprint requests to David J. Weiss, MD, Schwab Rehabilitation Hospital & Care Network/University of Chicago Pritzker, 1401 South California Boulevard, Chicago, IL 60608.

0 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

0003-9993/98/7909-4812$3.00/O

stated that 4 months earlier, he had a right great toe bunionec- tomy, and ambulated with axillary crutches for approximately 3 weeks afterward. He attributed his back pain to the crutch use. He had consulted two orthopedic surgeons in the previous 8 weeks and had been started on various nonsteroidal antiinflam- matory medications (NSAIDs) as well as prescribed physical therapy. There was no decrease in his pain level. The patient described the pain as a bandlike, constant dull ache between vertebral levels TlO and L2. The pain did not radiate, and he quantified the intensity as an 8 to 9 of 10 on a visual analog scale. His pain was exacerbated by prolonged sitting or standing and by twisting at the waist. Furthermore, the pain was worse at the end of the day and was occasionally severe enough to awaken him from sleep. He denied exacerbation of his symp- toms with Valsalva maneuvers. He noted that NSAIDs provided some relief from the pain. On a review of systems, the patient denied bowel or bladder dysfunction, chest pain, or abdominal pain. He was active physically, walking at least 1 to 2 miles per day, and following his home exercise physical therapy regimen of various lower extremity flexibility exercises. He was a nonsmoker and an occasional user of alcohol, ingesting approxi- mately one to two glasses of wine a week.

The patient was a well-developed, well-nourished man in no apparent distress. His vital signs were stable. The neuromuscu- lar examination showed normal sensation to light touch and pinprick. Manual muscle examination of the hip flexors, knee extensors, ankle dorsiflexors, and great toe extensors and flexors were normal. Hip abduction strength was graded as 3+/5, with a normal muscle firing pattern5 His hip extensor strength was also graded as 3 +/5, but with an abnormal muscle firing pattern5 His deep tendon reflexes were symmetric.

There was no palpable tenderness along the paraspinal muscles, spinous processes, the greater trochanters, sciatic notches, piriformis muscles, and the sacroiliac joints bilaterally. He had no pelvic obliquity deformities and no leg length discrepancy. He had bilateral hip flexor tightness in supine Thomas testing and was lacking 30” of flexibility in his hamstrings when tested in a supine position with his hips at 90” of flexion. Also, his iliotibial band and hip adductors lacked full flexibility bilaterally. On range of motion testing of the lumbar spine, he had full extension and was lacking approximately 25% of forward flexion and lateral flexion bilaterally. His lumbosacral lateral rotation was severely limited bilaterally by pain. All provocative tests (Gaenslen’s, Patrick’s, straight leg raises, piriformis, sacroiliac joint compression) were negative for reproduction of pain.

A tentative diagnosis was sacroiliac joint dysfunction second- ary to pelvic girdle muscle imbalances. This diagnosis was based on the history of a forced kinetic chain abnormality, that is, the use of axillary crutches after his bunionectomy, and his various physical findings. These findings included a lack of flexibility in his pelvic girdle postural muscles (hip flexors, hamstrings, hip adductors, iliotibial band)7 and weakness of certain pelvic girdle muscle groups (hip abductors, hip exten- sors). It was believed that this typical combination of decreased flexibility and weakness in critical muscles that crossed his sacroiliac joint resulted in sacroiliac joint dysfunction, which led to overuse of his erective spinae muscles and caused his mid-to-low back pain.

Arch Phys Med Rehabil Vol 79, September 1998

1138 LBP FROM A DUODENAL ULCER, Weiss

The patient was started on 25mg of nortriptyline at bedtime to improve his sleep. He requested that he continue his current NSAID, diclofenac sodium (Voltaren).2 Additionally, physical therapy was prescribed, and he was instructed on a home exercise program with a follow-up visit scheduled for 4 weeks. For the next 3 weeks, he experienced the same signs and symptoms, and without his physiatrist’s knowledge he in- creased his NSAID use. Four weeks after the initial office visit, he began to experience syncopal episodes and dyspnea on exertion. After an episode of coffee grounds emesis, he sought medical attention in a hospital emergency room. Pertinent findings on his admission physical examination included a pale complexion, orthostatic hypotension, an unremarkable abdomi- nal examination, and heme-positive stool. Coffee grounds were aspirated from his stomach with a nasogastric tube. The initial emergency room hemoglobin and hematocrit values were 3.0 and 9.6, respectively. He also had an elevated blood urea nitrogen level of 31 and a creatinine level of 0.9. Other initial laboratory studies, electrocardiogram, cardiac enzymes, electro- lytes, and coagulation studies were unremarkable.

Admitted urgently to the intensive care unit with a diagnosis of upper gastrointestinal bleed, he emergently received a transfusion of four units of packed red blood cells. An esophagogastroduodenoscopy showed a large duodenal ulcer in the duodenal bulb with associated inflammation of the entire pyloric channel. Also noted was a large crater with a visible blood vessel and evidence of recent bleeding. On the second day after admission, he underwent repair of the duodenal ulcer through vagotomy and pyloroplasty. His postoperative course was essentially unremarkable. He remained hemodynamically stable with a hemoglobin level of 10.5. He was slowly advanced to a regular diet and discharged to home on hospital day 8, postoperative day 7. His only discharge medication was omeprazole (Prilosec).6

The patient was seen in his physiatrist’s office approximately 3 weeks after hospital discharge. At that time, he stated that he felt markedly improved and that he had returned to a near normal activity level. He rated his pain as 2 to 3 of 10 without a change in the location of his symptoms, which he and his general surgeon attributed to a still healing ulcer. Physical examination was unchanged from initial evaluation, with the exception that pain was not elicited with lumbosacral range of motion testing. His lateral rotation had been limited by pain on his initial physiatry evaluation. He had normal sensation, deep tendon reflexes, and no palpable tenderness. Motor examination and muscle flexibility testing were unchanged. Also, provoca- tive tests were again nonpainful.

DISCUSSION The clinical picture of pain associated with a visceral

pathologic condition is complicated and has many components that may be both visceral and somatic in origin. Both visceral and somatic pain are associated with strong motor responses that may cause spasm, myofascial regional pain, and hyperalge- sia.6 The various theories of how a pathologic condition in a visceral structure causes referred somatic symptoms have been grouped into central or peripheral categories. The peripheral theorists have focused on a direct peripheral convergence of visceral and somatic afferent nerve endings; that is, the somatic peripheral pain afferent nerves are activated in some manner by various visceral fibers6 Conversely, the centralists have concen- trated on the spinal cord. One central theory focuses on visceral and somatic inputs into the spinal thalamic tract, the main pain tracts of the spinal cord. In this theory, the visceral afferent stimuli are misinterpreted by the spinal thalamic tract as being from the more common painful somatic stimuli. This has been

Arch Phys Med Rehabil Vol 79, September 1998

labeled “faulty localization.” Another central theory is that visceral afferent activation leads to an irritable focus in the spinal cord itself. This irritable focus is at multiple spinal cord levels that can be activated by many different stimuli, including nonpainful somatic stimuli. In 1995, Ness6 reviewed the literature on visceral pain and concluded that a combination of the central and peripheral theories is the best way to account for all of the symptoms that surround referred visceral pain.

In our case, the principal cause of the patient’s low back pain proved to be referred visceral pain from his duodenal ulcer. However, there is a question of whether his initial mid-to-low back pain could have been the result of sacroiliac joint dysfunction secondary to pelvic girdle muscle imbalances that were triggered by his crutch use. Thus, as the central theories hypothesize, the brain misinterpreted this new activity within his duodenal visceral neurons as being from the more com- monly experienced sacroiliac joint somatic stimuli, that is, “faulty localization.” Figure 1 highlights the visceral nervous system that makes this theory possible. In other words, our patient’s pain course could have come from a combination of sacroiliac joint dysfunction and his duodenal ulcer. The sacro- iliac joint dysfunction gave him his baseline pain level of 2 to 3

Fig 1. Visceral nervous system with highlighted pathologic path- way. (Copyright 1983, Novartis. Reprinted with permission from Ciba Collection of Medical Illustrations, Vol 1, illustrated by Frank H. Netter, MD. All rights reserved.)

LBP FROM A DUODENAL ULCER, Weiss 1139

out of 10 that he started and ended with. The visceral ulcer superimposed itself on this baseline pain level and added 6 to 7 points to his pain, giving him a level of 8 to 9 of 10 at its peak. With the resolution of his visceral pathologic condition, the pain level returned to its baseline of 2 to 3, that is, that being caused by sacroiliac joint dysfunction. This might explain why he continued to have some low back pain despite an operative cure of his visceral pathologic condition.

This case report reinforces the point that unusual causes such as peptic ulcer disease and abdominal aortic aneurysm should remain part of the differential diagnosis of mid-to-low back pain. These visceral referred pain symptoms are an excellent warning signal that, if acted on, can help prevent a true medical emergency.

References 1. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low

back pain. Spine 1995;20: 11-9.

2.

3.

4.

5.

6.

7.

a.

b.

Harris KM, Rosenbloom M. Images in clinical medicine: aortic intramural hematoma. N Engl J Med 1997;336:1878. Deathe AB. Hematometra as a cause of lumbar radiculopathy. Spine 1993;18:1920-1. Kiser TS, Mauldin CC, Grant R. Acute low back pain secondary to retroperitoneal hemorrhage in an elderly man. Arch Phys Med Rehabil 1997;78:664-5. Greenman PE. Principals of manual medicine. Baltimore (MD): Williams &Wilkins; 1996. Ness TJ. Historical and clinical perspectives of visceral pain. In: Gebhardt GF, editor. Progress in pain research and management, Vol 5. Seattle (WA): IASP Press; 1995. Singer KP. A new musculoskeletal assessment in a student popula- tion. J Orthop Sports Phys Ther 1986;8:34-40.

Suppliers CibaGeneva Pharmaceuticals, Ciba-Geigy Corporation, 556 Monks Avenue, Summit, NJ 07901. Astra Merck Inc., 725 Chesterbrook Boulevard, Wayne, PA 19087- 5677.

Arch Phys Med Rehabil Vol 79, September 1998