Don’t Shrug Off the Painful Shoulder: High-Risk Presentations Associated With a Common Chief Complaint
By James R. Hubler, MD, JD, Department of Emergency Medicine, University of Illinois, Peoria, Illinois; and James G. Zimmerly, MD, JD, MPH, LLD, (HON) Georgetown University Law Center, Washington, DC, and University of Maryland School of Medicine, Baltimore, MD.
The authors have provided an excellent review of the potential dangers when evaluating patients presenting with shoulder pain, particularly if the evaluation is perfunctory. In addition, the cases that they have selected for presentation reinforce some of the basic rules of emergency medicine (e.g., never rely on a previous doctor’s diagnosis [it might have been incorrect], take advantage of the opportunity that return visits provide to correct past mistakes, make sure all x-ray discrepancies are appropriately managed). These lessons, presented here in the context of the patient with shoulder pain, are applicable to all emergency department patients.
The differential diagnosis of the patient with shoulder pain is extensive. There are not a lot of teaching cases that I specifically recall from medical school, but one that has stuck in my mind was a patient with a presenting complaint of shoulder pain. The patient presented with isolated shoulder pain and, unfortunately for the patient, his ultimate diagnosis was lung cancer. He had Pancoast’s Syndrome, a syndrome resulting from local extension of an apical lung tumor to involve the eighth cervical and first and second thoracic nerves. These patients often present with shoulder pain, often radiating into the arm in an ulnar distribution.1
It is crucial to always remember that, just because a patient has a history of trauma to his or her shoulder, the shoulder pain may not be caused exclusively by the trauma. It may be that the trauma was merely incidental to the underlying cause of the shoulder pain. It is all too easy for busy emergency physicians to focus on the most obvious diagnosis and miss concurrent, underlying problems. True enough, patients usually have a single cause of their complaint. I still remember the words of my pathology professor in medical school: "No matter how much you have to push and squeeze, it’s got to fit just one disease."2 This rule, while applicable in the vast majority of cases, is not always true. There was a second part to the rule to cover the exceptions to the rule: "A person can have as many diseases as he damn well pleases."3
Introduction
Shoulder pain is a common chief complaint of patients presenting to the emergency department (ED). It has been reported that approximately 5% of all visits to general practitioners are for shoulder complaints, with the incidence increasing with age.4 The frequency of presentation to EDs is likely in the same general range. Because of the multiple serious medical problems that sometimes initially present in an atypical fashion with only shoulder pain as a symptom, a careful history and physical examination is always essential in evaluating the patient with shoulder pain. In those patients presenting with a clear history of trauma, diagnosis and treatment is usually relatively straightforward. However, in some cases, even when there is a history of trauma, the underlying diagnosis is not orthopedic and the complaint of shoulder pain may be symptomatic of a serious, even life-threatening illness. A broad differential diagnosis is always essential in the evaluation of the patient with shoulder pain in order to avoid missing the many possible, sometimes obscure, causes of shoulder pain. Because of the numerous serious conditions that may initially present with shoulder pain, the medicolegal risks in treating patients with shoulder pain are significant.
The emergency physician must recognize that, as a result of referred pain, cervical spine or abdominal pathology may initially present with shoulder pain. In addition, patients with myocardial infarction, aortic aneurysm, pulmonary embolus, pneumonia, subacute bacterial endocarditis, ruptured liver adenoma, various malignancies, as well as a number of other conditions may initially present with a complaint of shoulder pain. In some cases, shoulder pain will be their only symptom. Malpractice claims have resulted from missing each of these diagnoses in patients presenting with a complaint of shoulder pain.
The following cases illustrate the wide variety of conditions that may cause patients to present to their physician or the emergency department with a chief complaint of shoulder pain. Highlighted are some of the common life-threatening conditions that may present atypically with a complaint of isolated shoulder pain.
Intra-Abdominal Pathology
Shoulder pain can be a symptom of intra-abdominal pathology (e.g., bleeding). Because of diaphragmatic irritation, intra-abdominal bleeding may result in referred pain to the shoulder via the phrenic nerve.
Case #1. Schuler v Berger5 involved the postpartum care of a 24-year-old woman who developed left shoulder pain associated with back and abdominal pain in the immediate postpartum period. No diagnostic studies were ordered by her physician and she was treated with analgesics and sedatives. When the patient became hysterical and was screaming, the physician attributed it to postpartum psychosis.
When the patient’s husband called the defendant physician about his wife’s level of pain, he was told that the patient’s continued pain was due to "normal postpartum cramps as magnified by a low pain threshold."6 Unfortunately, within three hours of the husband’s call, the patient was found in profound shock with no pulse or blood pressure obtainable and with a distended abdomen. Resuscitation attempts were unsuccessful. An autopsy showed diverticulitis of the sigmoid colon with an acute rupture of a diverticulum resulting in peritonitis. The decedent’s estate was successful in its malpractice action against the treating physician.
This case represents a medicolegal disaster. The medical record should rarely reflect a patient’s low pain threshold. The evaluation and assessment should be limited to relevant subjective and objective findings from the history and physical examination. A psychiatric diagnosis in a patient with a complaint of pain is likely to be looked upon very unfavorably by jurors when the patient died of a non-psychiatric medical problem. It may appear that the physician has inappropriately discounted the patient’s complaint of pain. A presumptive psychiatric diagnosis may degrade the physician-patient relationship and, in retrospect, seem to reflect a physician’s lack of compassion. A diagnosis such as "non-specific abdominal pain" or "abdominal pain, etiology uncertain" may not have made this case a winner for the physician, but it certainly would have helped.
Commentary. The authors present an excellent example that illustrates several very important points. First, this is a case where the patient complained of shoulder pain, albeit with concurrent back and abdominal pain, and the diagnosis of peritonitis was missed. Second, the case demonstrates the folly of making a psychiatric diagnosis in a patient with signs and symptoms, while consistent with the psychiatric diagnosis, also entirely consistent with alternative, non-psychiatric diagnoses, without first adequately considering the possible non-psychiatric diagnoses and, where appropriate, ruling them out. In cases where the symptoms are consistent with both psychiatric and non-psychiatric diagnoses, the safe course is to make your psychiatric diagnosis a diagnosis of exclusion. Finally, the case provides a good example for the authors’ admonition to be careful with including gratuitous subjective comments in the medical record which, in retrospect, might appear judgmental and uncaring (e.g., "unkempt," "uncooperative," "poor pain threshold"). An "uncooperative" patient may not have been uncooperative due to a drug or psychiatric reason, as the treating physician might initially have thought at the time, but rather because of an overlooked, serious medical condition (e.g., pain, hypoxia.).
Case #2. In Madden v Linhardt,7 a 58-year-old woman died of complications following gallbladder surgery. Post-operatively she allegedly complained of shoulder pain. The defendant physicians felt that, because shoulder pain is common after gallbladder surgery, no serious condition existed. Without any investigation of the cause of her pain, they treated her in the postoperative period with narcotic analgesics. Her estate successfully argued that a retained common bile duct stone had been the cause of her postoperative right shoulder pain and her physicians’ failure to appropriately make this diagnosis proximately caused her death. After finding for the plaintiff, the jury awarded $2,812,000 in compensatory damages. This case illustrates a second type of abdominal pathology that may present initially with referred pain to the shoulder.
Other causes of referred shoulder pain include ruptured ectopic pregnancy, splenic injury (referred to as Kehr’s sign), ruptured abdominal aortic aneurysm, and ruptured adenoma of the liver.
Commentary. This case, like Schuler, is another excellent example of the risk of making a presumptive diagnosis (normal postoperative pain) and treating shoulder pain with analgesics without considering other more serious possible causes and, where appropriate, ruling them out. There is, of course, nothing wrong with giving narcotics for pain, in fact studies would suggest that emergency physicians are too conservative in their use of narcotics. However, prior to or in conjunction with the use of narcotics, a proper investigation of the possible causes of the patient’s pain is always necessary.
Case #3. In Kathleen Henry v the United States,8 a 26-year-old patient presented to a U.S. Army Hospital ED with a sole complaint of right shoulder pain of several days duration. She denied any history of trauma. Her only current medication was oral contraceptives. X-rays of her right shoulder were normal and she was discharged. Her mother, who was a nurse, insisted that she return to the hospital the next day because of persistent right shoulder pain. An internist staffing the ED found a pleural friction rub. That, coupled with the history of oral contraceptive use, caused him to suspect pulmonary embolus as the cause of her pain. A second internist agreed and IV heparin therapy was begun.
Shortly after admission, the second internist became concerned that the cause of the shoulder pain was located in the patient’s right upper abdomen and not her chest. He obtained a surgical consultation and the surgeon felt that a gallbladder etiology was very likely. The plan was to continue the heparin overnight, provide relief from the pain, and then further investigate the possibility of a gallbladder etiology the next day.
The patient suffered a cardiopulmonary arrest in her hospital room that evening. During the attempts at resuscitation her abdomen was noted to be increasingly distended. She was taken emergently to the operating room and, upon laparotomy, a massive amount of blood was found. Attempts at resuscitation were unsuccessful. Autopsy revealed a ruptured adenoma of the liver, a rare but recognized condition in patients taking oral contraceptives.
The heparin had exacerbated the hemorrhage into the adenoma, the original cause of the patient’s pain. The capsule of her liver that had initially contained the hemorrhage eventually ruptured and she exsanguinated. A malpractice suit was filed on behalf of the decedent. The case was scheduled for trial in the Federal District Court for the Northern District of Virginia when, at a pre-trial hearing, it was settled. At the time of this lawsuit, there had been only one published journal article associating adenomas of the liver with oral contraceptive pills. Since then the association has become widely recognized.9
Commentary. A good example of the potential risks of the use of anticoagulants in patients without a definitive diagnosis. In this time of cost-containment, when a patient presents at night, it is not uncommon to initiate heparin therapy before venous doppler studies are obtained in the morning. A related concern is the initiation of thrombolytics in patients who were thought to have AMI, but actually suffer from thoracic aortic dissection.
Pulmonary Embolism
Case #4.In a 1994 case, Matney vs Lowe,10 the plaintiff, after undergoing spinal fusion, developed shoulder and pleuritic right chest pain and fever of 101.4°F in the postoperative period. His evaluation included arterial blood gases, complete blood count, chest x-ray, and a ventilation-perfusion scan of the lungs. The chest x-ray was read as showing "vaguely defined areas of infiltration through the lung bases, suggesting an inflammatory etiology."11 The lung scan was read as "low probability for pulmonary embolus."12 The patient was also found to have a leukocytosis. The treating physician, relying on the interpretations of the chest x-ray and lung scan, made a diagnosis of pneumonia and treated the patient with intravenous antibiotics. The patient was subsequently discharged after four days of antibiotics. The day following his discharge he died at home and a subsequent autopsy found the cause of death to have been pulmonary embolism. There were multiple physician defendants in the case, all but one of whom settled prior to the trial. The remaining defendant, the surgeon, was found not liable.
Pulmonary emboli frequently present with atypical symptoms.13 Pulmonary emboli may occur concomitantly with other major illnesses such as pneumonia or congestive heart failure. The classic signs and symptoms of dyspnea, chest pain, cough, hemoptysis, tachycardia and tachypnea may not be present, or may be attributed to another illness. In this case, the patient’s signs and symptoms, while resulting from pulmonary embolism, were attributed, not unreasonably, to pneumonia. There was evidence admitted at trial that the lung scan may have been more correctly interpreted as indeterminate, rather than normal.
In a post-operative patient with dyspnea and pleuritic chest pain, the clinical suspicion of pulmonary embolus must always be very high. In this case, all but one of the defendant physicians settled prior to trial and, despite the fact that the jury found no cause of action against the remaining physician, juries are unpredictable and it is easy to imagine another jury finding for the plaintiff.
Commentary. If your clinical suspicion is high enough, a negative test, particularly a lung scan, should not completely outweigh your clinical judgment. It is always necessary to correlate the result of a test with what you feel is the pre-test likelihood that the condition is present. Evaluation of the patient with possible pulmonary embolism is excellently addressed in the PIOPED studies which should be familiar to all emergency physicians.
Neoplasm
Case #5.In Wilson v United States,14 a 50-year-old woman with a breast nodule was initially diagnosed as having fibrocystic disease. Five years later she returned to the same medical center with left shoulder pain. Over a six-month period her shoulder symptoms failed to improve and eventually, after several return visits, x-rays of her shoulders were obtained. The x-rays revealed lytic lesions of the left scapula and right clavicle. Despite this x-ray finding, the treating physician failed to diagnose metastatic cancer.
Four months later the patient was seen by another physician who, after reviewing her previous medical records and evaluating the patient, made the diagnosis of metastatic breast cancer. Unfortunately, at this stage only palliative care was indicated, and she died two years later. The patient’s daughter successfully brought an action for wrongful death.
Commentary. When appropriate, consider the possibility of metastatic disease in a patient with shoulder pain, including the possibility of a pathologic fracture when a fracture is present.
Case #6. In a similar case, Mays v United States,15 a woman presented with complaints of recent onset of cough and voice loss. On her initial visit, a chest x-ray was obtained that revealed a shaggy round lesion in the lower lobe of her left lung estimated to be four centimeters in size. The reviewing radiologist recommended follow-up x-rays in 7-10 days to exclude the possibility of carcinoma or organized infarct. The patient was never notified of this recommendation or advised to return for further evaluation and no follow-up testing was performed.
The patient returned to the same medical center two months later with a complaint of right shoulder and neck pain. At this time, x-rays were obtained but no action was taken regarding the previous chest x-ray or the previous recommendation for follow-up. One year later, the patient began experiencing fatigue and had to quit her job. She again returned to the same medical center and was found to have hypertension. At this time another chest x-ray was obtained that showed the same lesion seen on x-ray over one year earlier, now five and one-half centimeters in diameter. She underwent a left lower lobe lobectomy and radiation therapy for what was eventually found to be adenocarcinoma. She suffered multiple complications from radiation treatment, including radiation-induced myelitis, which left her a paraplegic. She died five years later at age 49 and, after a successful malpractice suit, the patient’s family was awarded $425,000 in damages.
This case demonstrates the importance of notifying patients with suspicious findings of the need for follow-up testing and ensuring that appropriate follow-up is obtained. This patient was never notified of the lung lesion. Then, when she returned to the same medical center two months later for evaluation of her neck and shoulder pain, the physicians compounded the original error by not noting the earlier chest x-ray finding and taking appropriate action. Most emergency department x-rays are "over-read" by a radiologist. The ED must have an effective system to handle x-ray discrepancies. If the patient has a primary care physician, both the patient and the physician should be notified of the findings and the need for additional follow-up evaluation. Those patients lacking a primary care physician should be contacted by telephone. If this is not possible, the patient may be notified by mail. Clearly, the more critical the possible problem, the greater the required diligence in contacting the patient. In emergencies, the police may be able to assist in locating and contacting a patient.
Commentary. The importance of a foolproof system for the follow-up of x-ray discrepancies cannot be overemphasized.
Case #7. In Swain v Garribrant,16 the patient, a prison inmate, complained to the prison physician of "knots under his left shoulder" for five months. The physician, Dr. Garribrant, allegedly diagnosed a "pulled muscle." Three other physicians also evaluated the patient for his shoulder pain during the six months prior to the patient being correctly diagnosed. The plaintiff argued that Dr. Garribrant was negligent in failing to recognize the symptoms of Hodgkin’s disease and allowing his disease to progress to an incurable stage in the interval between his first visit for evaluation of shoulder pain and the time when the diagnosis was ultimately made. The court, in this case, found for the physician in the subsequent lawsuit.
This case illustrates two important points. First, when patients return with the same complaint, a physician must broaden his or her differential diagnosis and consider diagnostic possibilities other than the original diagnosis. Second, physicians should not rely on the previous diagnosis of another physician. Repeat patients ("frequent flyers") are sometimes considered a burden by emergency physicians and their evaluation, as a result, is often compromised. In fact, the better practice is to consider "bounce backs" as opportunities to correct possible errors.
Commentary. No matter how good you believe the clinical skills of the previous physician to be, you owe it to the patient, as well as to yourself, to give the patient an unbiased evaluation. It is altogether too easy to allow the previous physician’s diagnosis to unduly influence your evaluation. As a result, you may blindly follow the path of the first physician, a path that may have taken a dangerous wrong turn. The first physician may have been wrong or missed the diagnosis. Disease evolves and the symptoms and signs may simply not have been sufficiently developed for the diagnosis to have been made at the time of the earlier visit. It is noteworthy that this case involved a civil rights claim brought under §1983 in which the burden of proof for the plaintiff was higher than in the "usual" malpractice case.
Case #8. In Ross v Hatchette,17 the patient was involved in an accident while driving a truck during the course of his employment. The patient received injuries to his right shoulder, neck and left knee and was hospitalized for four days. At the recommendation of his workers’ compensation attorney, he was seen by another physician at which time he complained of headaches and neck pain radiating to his left shoulder. He reported that he was improving and that his pain was not severe. The physician noted no atrophy to the left shoulder and range of motion was normal. Two months later the patient returned, now complaining of more severe pain in the left shoulder. Exam revealed a decreased range of motion. The physician again attributed the pain to the original neck injury. On the next visit, two weeks later, the physician noted that the patient was having extreme difficulty with his shoulder. X-rays of the shoulder were obtained and revealed a giant cell tumor of the head of the humerus.
The patient brought suit against the physician alleging failure to timely diagnose the tumor. In this case, the physician was found after trial not to have been negligent. Nevertheless, the case illustrates the potential problem of making a diagnosis and then failing to reconsider it when the patient returns for further evaluation.
Commentary. This case is an excellent example of the lesson that, not only should you not blindly trust the diagnosis made by a previous physician, you should not fail to reconsider your own previous diagnosis when the patient returns for further evaluation, particularly when the patient’s symptoms are worsening. It was certainly logical in this case to attribute the patient’s shoulder pain to his previous neck injury; it just wasn’t correct.
Case #9. In Ferrara v South Shore Orthopedic Associates,18 the patient presented to an orthopedic surgeon with a complaint of "pain in the right shoulder area." X-rays of the cervical spine, right clavicle and sternoclavicular joints were obtained in November, 1982 and were interpreted as showing "no evidence of fracture" and only "mild subluxation of the right sternoclavicular joint." The physician had also noted a mass in the patient’s right axilla. The patient was found seven weeks later to have "Stage III M1 adenocarcinoma of [the] lung metastatic to [the] anterior chest wall and right axillary lymph nodes." While the November, 1982 x-rays, obtained by the defendant physician, had been "lost" by the time of the litigation, other x-rays obtained in February 1983 reviewed by plaintiff’s expert were interpreted as showing signs of the metastatic cancer. Logically, the tumor must have been present on the November 1982 x-rays as well.
Commentary. The defendant physician ultimately prevailed in this case, based on the fact that, even if he had misread the x-rays and failed to make a timely diagnosis of the tumor at the time of the patient’s November visit, the patient’s diagnosis was only delayed for seven weeks and the prognosis would not have been different. Had the patient’s tumor been less advanced, or had the patient waited longer to consult another physician, the result might well have been different. It is always prudent to have patients return for re-evlauation if they are not better in a specific number of days or weeks.
Myocardial Infarction
Atypical presentations of myocardial infarction are common. As we all know, but sometimes forget, patients with myocardial ischemia may present initially with isolated shoulder pain. The emergency physician must always maintain a very low threshold for the consideration of unstable angina or myocardial infarction, particularly in a patient with shoulder discomfort. Failure to diagnose acute myocardial infarction remains the condition responsible for the largest amount of money paid out in emergency department malpractice claims year after year—approximately 30% of the total dollars paid out.
Case #10. In Johnson v Loeffler,19 et al, a 44-year-old man presented to the ED complaining of the sudden onset of shoulder and back pain after attempting to start a snowmobile. He had a past history of MI at the age of 37 and of angina at times since then. The patient presented to the ED because of his concern that the shoulder and back pain were symptoms of a cardiac problem.
The emergency physician took a history and performed a physical examination, ordered laboratory tests, and an ECG. He interpreted the ECG abnormalities that he identified as evidence of a previous myocardial infarction and made the diagnosis of back and shoulder strain, discharging the patient home with an offer of pain medication.
Two hours later, the patient’s girlfriend called the ED to report that the patient’s shoulder and back pain were increasing and that he needed more pain medication. A second emergency physician, who had not previously examined the patient, but was aware of the patient’s evaluation and assessment, prescribed Valium. The second physician never reviewed the patient’s chart, never talked to the patient, and did not advise the patient to come to the hospital for a reassessment. Approximately four and one-half hours after his release from the ED the patient collapsed in cardiac arrest. Attempts at resuscitation were unsuccessful. The autopsy revealed that the patient had severe coronary artery disease with a 95% occlusion of the left main coronary artery, a 70% occlusion of the left circumflex coronary artery, and a 95% occlusion of the right coronary artery.
The plaintiff’s expert, an emergency physician, testified that the standard of care was that the patient should have been admitted to the hospital and, in response to the girlfriend’s phone call, she should have been instructed to immediately return him to the emergency department for reassessment. The plaintiff’s cardiologist offered supportive testimony on the standard of care and testified that hospitalization would likely have saved the patient’s life. In addition, the plaintiff’s cardiologist testified that the patient was an excellent candidate for bypass surgery and that his life could have been extended at least 10 years.
Commentary. The authors provide another excellent example of the possible consequences of focusing too narrowly on a possible muskuloskeletal cause of shoulder pain. We all make mistakes. Therefore, never miss a chance to rectify an earlier mistake. Had the second physician talked to the patient and reviewed the chart, he or she likely would have directed the patient to immediately return (preferably by calling 911) and the result might have been different.
Case #11. In Gibides v Powell,20 a 28-year-old, six-month pregnant patient complained of pain and tingling in her arms and shoulders. Her physician attributed her symptoms to carpal tunnel syndrome. Eleven days after her clinic visit she suffered a myocardial infarction. Approximately eight months later, she suffered a second myocardial infarction and died. A wrongful death and medical malpractice action was brought against the initial physician who had made the diagnosis of carpal tunnel syndrome. The crux of the plaintiff’s claim was that the physician had failed to take an adequate history. The defendant argued that the patient complained only of pain in her wrists. Unfortunately, his medical records reflected only a diagnosis of carpal tunnel syndrome, without specific documentation of the history or physical examination.
Evidence was introduced that the patient had complained to her co-workers two weeks prior to her clinic visit of "chest pains, tingling in her arms and across her shoulders, shortness of breath and tightness in her chest." The jury was permitted to infer that she would have told the physician of these complaints had he taken an adequate history.
Commentary. Always support your diagnosis with adequate documentation in the medical record. If your diagnosis turns out to have been incorrect, adequate documentation of your thought process can at least make it appear logical.
Case #12. In Morales v United States,21 the patient, a 69-year-old male with a history of hypertension and aortic aneurysm, presented to the ED with complaints of neck and shoulder pain which had been present for two days. He was taking Motrin, which had afforded him some relief from the pain. The patient complained to family and triage personnel of pain radiating to his arms, numbness in the hands, and jaw pain. The treating physician elicited a history of pain only in the cervical area and both shoulders, aggravated by body movement and relieved with Motrin.
The physician performed, at best, what was described by the court as a "pro forma examination"—"the physician did not take the patient’s pulse, did not use the stethoscope to auscultate the patient, seemed to pay no attention to respiration rate, and, without the benefit of any further examination, determined that Morales-Gonzalez was not an emergency case."22 No ECG or other cardiac evaluation was done. The patient was discharged home and told to return the next day to the clinic for cervical spine x-rays. Later that night he was found dead by his wife. An autopsy revealed severe coronary disease with the cause of death determined to be myocardial infarction.
Commentary. While in retrospect we wonder how these types of cases can occur, there is unfortunately a never-ending stream of similar cases.
Cervical Spine Injury
Case #13.In Lasley v United States, an unreported case, the patient fell from a stepladder, landing on his head. He went to a U.S. Air Force hospital ED with the sole complaint of right shoulder pain. X-rays of his shoulder were normal, he was given medication for pain and discharged. The next day he returned to the same ED complaining of severe right shoulder pain. His x-rays from the previous day were reviewed [always a good idea] and the second physician agreed that there were no problems obvious on the radiographs. He increased Mr. Lasley’s pain medication and again discharged him. The patient was in so much discomfort that, instead of proceeding home, he visited a local chiropractor. The chiropractor obtained technically poor, but nevertheless adequate, cervical spine x-rays from which a fracture-dislocation of C-6/C-7 could be diagnosed. The C-6 vertebral body was displaced approximately one centimeter anterior on C-7.
Risk Management Tips |
| 1. A detailed history and directed physical examination are always essential. |
| 2. Always remember that patients with myocardial infarction, pulmonary embolus and other life-threatening disorders may initially present with atypical symptoms (e.g., isolated shoulder pain). |
| 3. Be cautious of making a musculoskeletal diagnosis in patients with shoulder pain if you are unable to reproduce the pain. |
| 4. The presence of reproducible pain in the patient with a complaint of shoulder pain does not, however, categorically rule out serious pathology. Remember that, while the shoulder "sprain" may have caused the patient to come to the doctor, there may be a more serious, occult problem. |
| 5. Always consider a cardiac etiology in patients with shoulder pain and evaluate those patients appropriately. While younger patients may be less likely to have coronary disease, they are not immune. |
| 6. Close follow up is always necessary in patients with unusual symptoms or unclear diagnoses. Contact with the primary care physician will help to ensure follow up and continuity of care. |
| 7. Patients with repeat visits to the emergency department and/or other health care providers may require more extensive evaluation and testing. A return visit should always trigger a concern that the initial diagnosis was incorrect and should be treated as an opportunity to correct it. Always consider alternative diagnoses. |
| 8. Broaden your differential diagnosis when patients return. |
| 9. Never rely to a previous physician's diagnosis. Always try to evaluate return visits with an open mind. |
| 10. A low threshold for x-rays, particularly in the elderly, may reveal occult malignancy or fracture. Editor's note: Cost containment and risk management advice is not always the same—a balance is required. |
| 11. If the signs and symptoms are consistent with both a psychiatric and non-psychiatric diagnosis, make the psychiatric diagnosis a diagnosis of exclusion. |
| 12. Avoid gratuitous statements in the medical record ("low pain threshold." "uncooperative," "unkempt," etc.). If it turns out you missed the diagnosis, such statements can make you appear uncaring or unsympathetic. |
| 13. A foolproof system for the follow-up of x-ray discrepancies is essential. |
| 14. Always instruct patients to return in a specified number of days if they are not improved. |
| 15. Never give phone advice to patients without reviewing the record of their previous visits. The safest advice is always to instruct the patient to return for a recheck. |
| 16. Support your diagnosis with adequate documentation in the medical record. If you missed the diagnosis, adequate documentation of your exam and thought process can at least make your conclusion appear logical. |
The chiropractor protected the patient’s neck and sent him by ambulance back to the Air Force hospital ED.
Commentary. The patient and the defendants were both fortunate in this case because the patient experienced no neurologic deficits and, therefore, his malpractice claim was limited to damages for one extra day of pain and suffering.
Case #14. Shoulder pain can be the predominant location of referred pain from injuries to the cervical spine. In Parson v Keys and Wilkes Regional Medical Center, another unreported case, a 34-year-old male was the driver in a serious single-car MVA. Upon arrival at the ED he complained of left shoulder pain as well as neck and low back pain. The patient had tenderness in his lower cervical spine, as well as his left shoulder. He had a full range of motion in all extremities. Because of technically inadequate cervical spine x-rays, as well as physician misreading, a fracture-subluxation of C-7 was missed. The patient was incorrectly diagnosed as having a "cervical neck strain."
The fracture was difficult to identify on plain films and on re-evaluation a CT scan was required to definitively identify it. The patient unfortunately developed a partial paralysis, as a result of the injury, prior to his transfer and definitive treatment at a regional trauma center. The case against both the attending surgeon and the community hospital was settled prior to trial.
In this case the patient’s midline tenderness on palpation of the cervical spine required a complete cervical spine visualization including the C-7/T-1 junction. Most emergency physicians understand that patients who present with midline cervical pain, weakness, paralysis, abnormal sensation, altered level of consciousness, or intoxication in the face of possible trauma must receive a complete cervical spine radiologic examination.23 The threshold for complete radiologic evaluation should be low in patients with significant distracting injury and also in the intoxicated or confused patient.24 Remember also that a complaint of abnormal sensation localized to a single nerve root level may be the only clinical finding associated with a potentially unstable spinal cord injury.25
Commentary. Never accept technically deficient x-rays. This is particularly true when you have a significant pre-test likelihood, based on history and exam, that there might be a fracture.
Other Causes of Referred Pain to the Shoulder
On February 18, 1992, 28-year-old Sharon Bradley presented to the ED with a history that included left shoulder pain, fever and chills.26 Her history included septic arthritis, chronic osteomyelitis of her right foot, and subacute bacterial endocarditis. Her vital signs were normal. A fairly complete evaluation, including CBC, blood culture, urinalysis, ABGs, chest and shoulder x-rays, was normal with the exception of the blood culture, results of which, of course, were not available on the day of examination. The working diagnosis was pleurisy, and the patient was prescribed Percocet and directed to return the next day.
She returned as directed the following day feeling much worse. After evaluation by the second emergency physician, the internal medicine resident and an orthopedist, she was admitted. She subsequently died as a result of Staphylococcus aureus endocarditis.
Commentary. While cases such as this may be rare, they are still possible.
Conclusion
Most commonly, complaints of shoulder pain represent symptoms of routine musculoskeletal conditions. However, physicians must always remember that shoulder pain may be a harbinger of serious, sometimes life-threatening, disease. If the physician fails to appropriately consider such disorders, significant medicolegal risk may be incurred. As in other high-risk situations, there is no substitute for a high index of suspicion, thorough evaluation, and careful follow-up.
References
1. Fauci AS, et al, eds. Harrison’s Principles of Internal. 14th ed. 1998.
2. Paul Gikas, MD, Professor of Pathology, University of Michigan Medical School (1978).
3. Id.
4. The Johns Hopkins Medical Letter, Sept. 1998.
5. 275 F.Supp. 120 (E.D. Pa. 1967).
6. Id. at 122.
7. Unreported Maryland case, Estate of Madden v Linhardt, MD and Anne Arundel Medical Center, C 96-18963.
8. Unreported case, (N.D. Va 1975).
9. Baum JN, et al. Possible association between benign adenomas and oral contraceptives. Lancet 1973:926; Klatskin B. Hepatic tumors: Possible relationship to use of oral contraceptives. Gastroenterology 1977:386; Edmondson HA , et al. Liver cell adenoma associated with use of oral contraceptives. N Engl J Med 1978:470.
10. 191 W.Va. 220, 444 S.E.2d 730 (1994).
11. Id. at 221.
12. Id.
13. Fisher WT, et al. Atypical presentations of pulmonary embolism. Ann Emerg Med 1990;19:1429; Plowman K and Jones JS. Diagnosis and treatment of pulmonary embolism in the elderly. Ann Emerg Med 1991;20:463.
14. 637 F.Supp. 669 (E.D. Va. 1986).
15. 608 F.Supp. 1476 (D. Col. 1985), rev’d and remanded, 806 F.2d 976 (10th Cir. 1986) (reversed as to Colorado’s interpretation of the collateral source rule).
16. 354 F. Supp. 631 (E.D. N.C. 1973).
17. 251 So. 2d 820 (La. Ct. App. 1971).
18. 178 A.D.2d 364 (Sup Ct. N.Y.App. Div. 1991).
19. Unreported Minnesota case, Dawn Johnson v. Dale Loeffler, Suzan Vitalis and Chicago Lakes Hospital Chicago County (Noted in Sullivan D, Zalensi R. Missed myocardial infarction: Minimizing the risk. ED Legal Letter 1996;7:6.
20. 682 N.Y.S. 2d 771 (N.Y. App. Div. 1998).
21. 642 F.Supp. 269 (D. P.R. 1986).
22. Id. at 271.
23. Neifeld G, Keene J, Hevesy G. Cervical injury in head trauma. J. Emerg Med 1988;6:203.
24. Walter J, Doris PE, Shaffer MA. Clinical presentation of patients with acute cervical spine injury. Ann Emerg Med 1984;13:512.
25. Maroon JC. "Burning Hands" in football spinal cord injuries. JAMA 1977;238:2049 (1977).
26. Bradley v. Spectrum Emergency Care, Inc. et al.
Physician CME Questions:
9. The common complaint of shoulder pain can present a medicolegal risk because the underlying condition may be:
a. myocardial infarction.
b. neoplastic disease.
c. pulmonary embolism.
d. Any of the above
10. Rupture of an abdominal organ resulting in shoulder pain, is easy to diagnose because the history will always include a history of trauma.
a. True
b. False
11. Which of the following conditions is never associated with an atypical presentation?
a. Myocardial infarction
b. Pulmonary embolism
c. Both
d. Neither
12. Complaints of shoulder pain account for the what percentage of visits to primary care practitioners?
a. 2%
b. 5%
c. 9%
d. 12%
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content