To investigate epidemiology of acute non-traumatic back pain using modern diagnostic methods in patients who visited an emergency room.
To investigate epidemiology of acute non-traumatic back pain using modern diagnostic methods in patients who visited an emergency room.
The medical charts were retrospectively reviewed for all patients with back pain who were treated in our hospital. In addition, the patients were divided into two groups based on whether they were treated at the hospital or as outpatients.
There were 95 patients with non-traumatic acute back pain. Leading cause of back pain was ureterolithiasis (53 cases), followed by pyelonephritis (10), orthopedic disease including two cases of purulent spondylitis (24), aortic disease (3), pancreatitis (1), renal bleeding (1), adrenal bleeding (1), psoas abscess (1), and torsion of an ovarian tumor (1). All cases of pyelonephritis, aortic disease, purulent spondylitis, renal bleeding, adrenal bleeding, psoas abscess, and torsion of an ovarian tumor were treated in admission. Using a multiple logistic regression analysis, blood pressure, age, and body temperature were the only factors that were independently associated with whether the patient was admitted or treated as an outpatient.
This study showed that urological diseases are the most common cause of back pain in patients who visit the emergency room, followed by orthopedic disease. Older age, low blood pressure, and high body temperature were independently associated with the decision to admit the patient who might have lethal disease.
Many people with back pain visit the emergency room (ER). The most common cause of back pain is orthopedic disease, followed by urological disorders such as ureterolithiasis or pyelonephritis. Rarely, patients with aortic disease, angina pectoralis, pulmonary embolism, pancreatic, esophageal, duodenal or adrenal disease, spinal tumors, and infections or retroperitoneal fibrosis visit the ER complaining of back pain. Most of these causes of back pain can be lethal. Radiologists and physicians in other fields have reported the usefulness of computed tomography (CT) and magnetic resonance imaging (MRI) to obtain a correct diagnosis of back pain in addition to taking a comprehensive medical history of the present illness and carrying out a physical examination. However, there have been only review reports concerning back pain from the standpoint of ER physicians, and no epidemiological data about the analysis of back pain in the ER using these methods. Hence, we carried out a retrospective study to analyze the patients with back pain in our hospital.
This retrospective study was approved by our institutional review board. Juntendo Hospital is located in Tokyo and has 1020 beds. The ER in Juntendo University Hospital is classified as a primary and secondary medical facility from the standpoint of emergency care. In Japan, primary emergency medical facilities receive patients that are not in serious condition, who can walk in for treatment and do not need hospitalization, such as patients with flu, diarrhea, or simple contusions. Secondary medical facilities receive patients that cannot walk on their own and need hospitalization, but are not in a critical condition, such as those with pneumonia, appendicitis, or fractures. The ER in this hospital treats patients with moderate critical illnesses or trauma who are transported by ambulance or arrive by themselves, and also treats hospitalized patients who require emergency treatment, who are mainly treated by staff in the Department of Emergency and Disaster Medicine. The ER in this hospital also treats patients with mild illness, who are mainly treated by staff members of the Department of General Medicine.
The medical charts were retrospectively reviewed for all patients who listed back pain as one of their reasons for visiting the ER, and who were treated by staff in the Department of Emergency and Disaster Medicine at Juntendo Hospital between January 2011 and March 2013. The exclusion criteria were back pain induced by obvious trauma, such as a traffic accident, fall, or violence, or due to a skin infection. Individual medical records were reviewed with regard to the following data: sex; age; use of an ambulance; vital signs on arrival (systolic blood pressure, heart rate, Glasgow Coma Scale score, body temperature); the type of examination(s) carried out (blood analysis, urinalysis, X-ray, CT, MRI); the usefulness of the examination to obtain a correct diagnosis; the admission status of the patient; the final diagnosis; the mortality rate. In addition, the subjects were divided into two groups based on whether they were treated in the hospital (admission group) or as outpatients (outpatient group). The variables mentioned above were investigated in these two groups for comparison purposes.
The statistical analyses were carried out using unpaired Student's t-tests and the χ2 analysis. A P-value <0.05 was considered to indicate a statistically significant difference. All data are presented as the means ± standard error. A multivariate analysis using a logistic regression analysis was used to evaluate the independent factors regarding the difference between admission and outpatient data. The variables included in the multivariate analysis were those with significance levels of P < 0.05 based on the univariate analysis. The odds ratios and the corresponding 95% confidence intervals were also calculated. For the statistical analysis, StatView Version 5.0 (SAS Institute, Cary, NC, USA) was used.
During the study period, there were 3055 patients who visited the ER for the first time. Among them, 183 had back pain. Of these, 85 cases with trauma and 3 cases of skin infection were excluded from the study. The remaining 95 patients were included as the study subjects. Table 1 shows the background of the subjects.
|Diagnosis||Number of patients||Rate (%)||Ambulance||X-ray†||CT exam.†||Blood exam.†||Urine exam.†||Admission|
|(as a complication of ureterolithiasis, 1)|
|L. disc herniation||6||6.3||2||0/4||2/2||0/1||0||0|
|L. compression fracture||2||2.1||2||2/2||1/1||0||0||0|
|Spinal canal stenosis||2||2.1||1||0||1/1||0/2||0||0|
|Ruptured abdominal aorta||1||1.0||1||0/1||1/1||0/1||0/1||1|
|Torsion of an ovarian tumor||1||1.0||1||0||1/1||0/1||0/1||1|
Concerning the diagnostic methods, two cases were evaluated based on previous radiological data. At our hospital, CT examinations were frequently used (74/95, 77.8% of cases) as a diagnostic method. An emergency MRI was carried out in only two cases. One case had purulent spondylitis and the other had torsion of an ovarian dermoid tumor confirmed by an emergency operation. The MRI was not helpful for obtaining the correct diagnosis in either of these cases (Figs 1, 2). In the case with purulent spondylitis, inflammatory changes were obtained by a repeated MRI 18 days later. One case with an aortic dissection was misdiagnosed to have orthopedic disease based on a measurement of the blood pressure at the extremities, a chest X-ray, blood and urine analyses, and an electrocardiogram. This case revisited our hospital complaining of chest and continuing back pain, and enhanced truncal CT carried on the same day revealed the aortic dissection with increase of d-dimer (Fig. 3).
In the 19 patients who were admitted to the hospital, the majority (10 patients) had pyelonephritis. The presence of inflammatory changes in the blood and urine examinations, and a peri-renal dirty fat sign on the CT scan were helpful for obtaining the correct diagnosis in these patients. One case of adrenal bleeding was found to be due to a metastatic lung tumor (Fig. 4). One case of renal bleeding was from a right renal angiomyolipoma in a patient with tuberous sclerosis (Fig. 5). Two cases of purulent spondylitis were diagnosed by repeated MRI.
The two mortality cases consisted of one patient with a ruptured abdominal aorta and one with a psoas abscess. The patient with a ruptured abdominal aortic aneurysm died of uncontrollable hemorrhagic shock. The case with a psoas abscess did not undergo drainage because her relative did not agree with an invasive approach, and the infusion of antibiotics alone failed to control the infection.
The results of the analysis of the admission and outpatient groups are shown in Table 2. One patient with an aortic dissection that was initially misdiagnosed was admitted 2 days later, so that this case was classified into the admission group. No significant differences were observed between the two groups with regard to ambulance use or the usefulness of radiological studies. The average age, ratio of females, pulse rate, white blood cell count, and mortality rate in the admission group were significantly higher than those in the outpatient group. The average blood pressure and level of consciousness in the admission group were lower than those in the outpatient group.
|Admission (n = 19)||Outpatient (n = 76)||P-value|
|Age||66.2 ± 4.0||45.1 ± 1.6||<0.0001|
|Rate of ambulance use||13 (68.5%)||46 (30.0%)||n.s.|
|Glasgow Coma Scale score||14.7 ± 0.1||14.9 ± 0.0||<0.01|
|Systolic blood pressure (mmHg)||111.7 ± 4.8||136.2 ± 2.1||<0.0001|
|Heart rate (/minute)||91.4 ± 4.5||78.0 ± 1.9||<0.01|
|Temperature (°C)||37.1 ± 0.3||36.3 ± 0.1||<0.001|
|White blood cell count (/mm3)||n = 19||n = 57|
|1208 ± 150||8094 ± 323||<0.001|
|Usefulness of CT||n = 15||n = 59|
|13 (86.6%)||50 (86.2%)||n.s.|
|Usefulness of X-ray||n = 3||n = 21|
|0 (0.0%)||4 (19.0%)||n.s.|
|(compression fracture, 2 cases; spondylosis, 1 case; ureterolithiasis, 1 case)|
|Mortality rate||2 (10.5%)||0 (0.0%)||<0.05|
Using a multiple logistic regression analysis, we found that blood pressure (odds ratio [OR], 0.92; 95% confidence interval [CI], 0.87–0.97; P = 0.004), age (OR, 1.08; 95% CI, 1.01–1.15; P = 0.01), and body temperature (OR, 1.13; 95% CI, 1.02–1.26; P = 0.01) were the only factors that were independently associated with whether the patient was admitted or treated as an outpatient.
This study shows the frequency of the diagnosis of different causes of back pain in patients who visited the ER, as determined using modern diagnostic methods. The study indicated that urological diseases are the most common cause of back pain in patients who visit the ER, followed by orthopedic disease. In addition, older age, low blood pressure, and a high body temperature were independently associated with the decision to admit the patient who might have lethal diseases.
There have been a number of reports concerning low back pain syndrome or spondyloarthropathy. Non-specific low back pain is not associated with neurological symptoms or signs and is often a result of simple soft tissue disorders. Low back pain can arise from the vertebral column, surrounding muscles, tendons, ligaments, and fascia due to the stretching, tearing, or contusion of these tissues. Most cases with these types of pain resolved spontaneously. Our study included patients with back pain complicating other symptoms and signs, such as abdominal pain or flank pain. In addition, cases with trauma were excluded from the present study, so that urological diseases (especially pyelonephritis) were the most frequent cause of back pain in our study.
In this study, there was no mortality in the outpatient group, so the decision regarding whether to admit the patient or treat them as an outpatient was considered to be correct, excluding one misdiagnosed case with an aortic dissection. Older age, low blood pressure, and high body temperature were independently associated with the decision to admit the patient. Older age is known to be correlated with immune suppression, the progression of atherosclerosis, and the occurrence of cancer, so older age was an important factor that led ER physicians to recommend admission. As the average age of the admission group was 66 years and the World Health Organization has defined an older person as being either 65 or >60 years, the suitable definition of older age may therefore be over 60 years of age. A high body temperature reflected a high frequency of inflammatory diseases, such as pyelonephritis, purulent spondylitis, and a psoas abscess in the admission group. Low blood pressure reflected the presence of septic shock or retroperitoneal hemorrhagic shock. Accordingly, if a patient with back pain is over 60 years of age and has a fever and/or low blood pressure, then aggressive examinations, including CT and biochemical studies, should be considered, similar to other complaints with these symptoms that would require detailed examination and diagnosis.
The main interest of the physician is a diagnostic pitfall of back pain, or how the physician can avoid misdiagnosis. Generally, many patients with back pain have non-life-threatening etiologies and recover within 4–6 weeks. Because it is a common syndrome with a generally benign origin, the examiner may overlook a small percentage of markers of serious life-threatening disease. Clinical suspicion for these diagnoses begins with a thorough history and physical examination, following appropriate laboratory studies and diagnostic imaging based on the suspected etiology. Some reports empirically suggest the red flag theory for determining life-threatening back pain. Examples of red flags in patient history are gradual onset, age under 20 years or over 50 years, thoracic back pain, pain lasting longer than 6 weeks, history of trauma, fever, unintentional weight loss, pain worse with recumbency, pain worse at night, history of malignancy, recent procedure known to cause bacteremia, or history of intravenous drug use. Examples of red flags in physical examination are fever, blood pressure abnormality, pale, pulsatile abdominal mass, pulse amplitude differentials, spinous process tenderness, focal neurologic signs, or acute urinary retention. However, red flags have very high false-positive rates, indicating that, when used in isolation, they have little diagnostic value in the primary care setting. In addition, the theory of red flag management is insufficient evidence to provide recommendations regarding their diagnostic accuracy or usefulness. Accordingly, there is no gold standard method to avoid a misdiagnosis of back pain. Carrying out all laboratory, radiological including CT/MRI, physiological, and ultrasound examinations for all patients with back pain may reduce the misdiagnosis of a small percentage of life-threatening diseases, however, this strategy is not considered to be cost-effective. Our findings might assist ER physicians in selecting patients who require further examination. However, even considering the three significant factors revealed in our study (high age, fever, hypotension), life-threatening conditions can unfortunately be missed, for example, cases with younger age, normal body temperature, and normotension on arrival (Figs 2, 5).
In this study, CT was frequently used to assist the differential diagnosis of back pain. This might be because: (i) CT is commonly used in Japan; (ii) CT has higher spatial resolution compared to plain radiographs; (iii) the medical expenses of outpatients are part of a piece-work system in Japan, so that CT examinations are financially beneficial to the hospital; (iv) misdiagnosis without CT results carries an increased risk of lawsuits. However, CT is associated with disadvantages, including exposure to radiation, so further prospective, large-scale studies are warranted to determine the efficacy of CT and to try and establish a gold standard method for evaluating back pain in the emergency setting.
In this study, urological diseases were the most common cause of back pain in patients who visited the emergency room, followed by orthopedic disease. Older age, low blood pressure, and high body temperature were independently associated with the decision to admit patients thought to have potentially lethal disease.