Misdiagnosis of ruptured abdominal aortic aneurysms William A. Marston, MD, Richard Ahlquist, MD, George Johnson, Jr., MD, and Anthony A. Meyer, MD, Phi), Chapel Hill, N.C., San Francisco, Calif., and
Spokane, Wash. Ruptured abdominal aortic aneurysm is a surgical emergency with a high mortality rate even when diagnosed and repaired immediately. We retrospectivelyreviewed 152 cases of ruptured abdominal aneurysms to identify the incidence of misdiagnosis leading to a delay in treatment, the most frequent misdiagnoses, and the outcome in this group of patients. Forty-six (30%) were initially misdiagnosed. The most common misdiagnoses were renal colic, diverticulitis, and gastrointestinal hemorrhage. The most common initial physical findings in misdiagnosed patients were abdominal pain (70%), shock (57%), and back pain (50%). A pulsatile abdominal mass was found in only 26% of misdiagnosed patients versus 72% of patients correctly diagnosed (p < 0.005). Misdiagnosed ruptured abdominal aneurysm had a 44% mortality rate, which was not significantly different from patients correctly diagnosed (58%,p = 0.34). The lack of difference in mortality rates is most likely due to preselection of those miscliagnosed patients who were able to withstand the delay in diagnosis and survive to surgical treatment. The 30% incidence of misdiagnosis in this series suggests that it is frequently a difficult diagnosis to make and must be considered in elderly patients, especially men, who are admitted with abdominal pain and/or back pain. (J VAse SURG 1992;16:17-22.) Rupture of abdominal aortic aneurysm (RAAA) is a catastrophic event that was a uniformly fatal complication of AAA until the 1950s, when emergent repair of RAAA first became possible. Results improved significantly in the next 15 years as a result of improvements in surgical, anesthetic, and postoperative management. However, the mortality rate remains quite high, reported at 1596 to 7096 in various series, and has not improved in the last 2 decades.l"z Early diagnosis is believed to be important in improving the outcome of RAAA, Several investigators examining the causes of death have identified an increased mortality rate for persons who are incorrectly diagnosed on admission, s'4,6 The frequency of misdiagnosis of RAAA is not well known nor are the medical diagnoses that are commonly confused with RAAA. In this article reports of RAAAs from three institutions were reviewed to
From the Departments of Surgery, the Universityof North Carolina Hospitals, Universityof California-San Francisco, and Spokane. Presentedat the North CarolinaChapterof the AmericanCollege of Surgeons,Asheville,N.C., July 14, 1991. Reprint requests: AnthonyA. Meyer,MD, PhD, Universityof North CarolinaDepartmentof Surgery,164 Burnett-Womack, Campus Box 7210, ChapelHill, NC 27599-7210. 24/1/34344
identify the incidence of misdiagnosis and characteristics of patients whose RAAAs are misdiagnosed. METHODS Patients. One hundred fifty-two patients from three institutions who sustained RAAAs between 1974 and 1989 were diagnosed either in a referring emergency department or at the regional referral center. Cases in which the rupture occurred during elective AAA repair were excluded, as were patients with no evidence of retroperitoneal or free bleeding at operation. The three institutions involved were the University of North Carolina Hospitals, a vascular referral center (58 cases); San Francisco General Hospital, a municipal teaching hospital (47 cases); and a community hospital practice in Spokane, Wash. (47 cases). Record review. Emergency depamnent records of both the initial and referral centers were reviewed for each patient to identify initial symptoms, hemodynamic stability, initial diagnosis, time until diagnosis of RAAA, and diagnostic studies performed. Misdiagnosis of RAAA was defined as a delay in diagnosis of at least 6 hours or an initial primary diagnosis other than RAAA. Two of the following three symptoms were required to diagnose shock: blood pressure less than 70 mm Hg, pulse greater
17
Journalof VASCULAR SURGERY
18 Alarston et al.
Table I. Findings on initial examination Diagnosis
No. of patients
Shock
Bkpn
Abdpn
Mass
Avg delay
Mortality rate
Renal colic Diverticulitis GI hemorrhage Acute MI Back pain MVA Sepsis Other GI Other/no Dx Total
11 6 6 4 4 3 3 3 6 46
6 3 3 2 2 3 3 3 1 26
9 2 0 2 4 2 0 2 2 23
9 6 5 2 1 2 1 3 3 32
1 1 2 2 0 2 2 0 2 12
15 hr 79 hr 17 hr 13 hr 18 hr 1.5 hr 26 hr 4 hr 18 hr 24.4
5 (45%) 1 (17%) 3 (50%) 2 (50%) 2 (50%) 3 (100%) 2 (67%) 1 (33%) 1 (17%) 20 (44%)
Bkpn, Back pain; Abdpn, abdominal pain; Mass, pulsatile mass on first examination; Avg delay, average delay in diagnosis; A4/, myocardial infarction; MVA, motor vehicle accident; G/, gastrointestinal; Dx, diagnosis.
than 125 beats per minute, or poor peripheral perfusion. For those patients identified with misdiagnosis of RAAA, hospital records were reviewed to determine operative characteristics and outcome. Mortality was defined as death before discharge from the hospital or within 30 days of surgery. Statistics. Differences between groups were tested by chi-square analysis by means of a microcomputer, with a p value of _ 8 hr (n = 20)
p value
22 (84.6%) 11 (42.3%) 16 (62%)
4 (20%) 3 (15%) 4 (20%)
0.015 0.13 0.07
series, and is reported to be increased by delay in operative treatment. 2.s We examined 46 cases of misdiagnosis to identify factors leading to misdiagnosis and to evaluate the effect of delay in treatment on outcome. Patients admitted with RAAA are typically elderly men, with 81% being 64 years of age or older. The symptoms most commonly found in this review of 152 cases were abdominal and back pain, at 76% and 54%, respectively. Both were present in only 42%, and the classic diagnostic triad of abdominal pain, back pain, and pulsatile mass was found in only 26% of patients. In those patients who were misdiagnosed, the classic diagnostic triad was found in only 9% of cases. Misdiagnosis was more frequent in those patients admitted initially to a referring emergency depat-ti~tent than in those admitted initially to a tertiary care center. When the frequency of symptoms was compared in patients correctly diagnosed versus those misdiagnosed, no significant difference was found in the incidence of abdominal or back pain. The incidence of shock was slightly higher in those correctly diagnosed. A previously known AAA was infrequent in both groups. However, the detection ofa pulsatile mass was much more frequent in those correctly diagnosed, with only one fourth of the misdiagnosed patients having a pulsatile mass on initial examination. Another 25% of misdiagnosed patients were found to have a mass on a later examination, usually performed by a surgical consultant. A pulsatile mass may not be identifiable in many patients. Obesity often obscures this finding, and many patients may
have enough bleeding in the retroperitoneum to cause the whole abdomen to have a distended, doughy feel. Furthermore, profoundly hypotensive patients may not generate a systolic pressure high enough to palpate a pulsatile mass. Only 14% of 88 patients with a pulsatile mass on initial physical examination were misdiagnosed, whereas 53% of the 64 patients with no pulsatile mass were misdiagnosed (p < 0.0003). In patients whose aneurysm did rupture, the presence ofa pulsatile mass was the most important finding aiding correct and rapid diagnosis, emphasizing the need for a prompt abdominal examination in all elderly male patients, by a surgical consultant if possible. The erroneous diagnoses most frequently made were those associated with abdominal pain, shock, and back pain, the symptoms commonly seen in this series. Diagnoses of intraabdominal processes, primarily sigmoid diverticulitis and gastrointestinal hemorrhage were made in nearly one third of cases. The incidence of gastrointestinal hemorrhage as a diagnostic error is partially due to the association of aortoenteric fistula with rupture in three patients in this study. Patients misdiagnosed with diverticulitis generally had long delays in diagnosis, usually until an abdominal CT scan was obtained when their symptoms did not improve on conservative therapy. These patients did well as a group, probably because of their hemodynamic stability and their improved volume status after several days of intravenous hydration before correct diagnosis of R A A . However, patients with RAAA unresponsive to such limited resuscitation may die with this incorrect diagnosis, and the true cause of death may never be known. Renal colic was also a frequent misdiagnosis, with most patients having severe back pain and two thirds having both back and abdominal pain. More than 50% had a hypotensive episode during the initial evaluation, but most of these were transient events, diagnosed as a vagal episode or a result of developing sepsis. In some cases the diagnosis was not made until
Volume 16 Number 1 July 1992
a second, more severe hypotensive episode occurred, suggesting that a more acute process was developing. Acute myocardial infarction, lower back pain, sepsis, and blunt motor vehicle trauma were the primary diagnoses in several cases each. Patients with RAAA may have myocardial ischemia, with characteristic changes on an electrocardiogram, probably caused by hypotension in the patient with diffuse atherosclerotic disease, which involved the coronary arteries as well as the aorta. Three of the four patients misdiagnosed with lower back pain had no abdominal examination documented on their initial examination record. In patients admitted with presumed blunt trauma after a motor vehicle accident, the degree of vehicular damage was mild in comparison with their symptoms. All three were in shock, but none had severe long bone fractures, significant chest injuries, or other evidence of major injury. General observations made from reviewing these patients' presentations are the following: (1) In elderly patients, especially white men, the possibility of symptomatic or RAAA must be considered in all patients who are admitted with any symptoms related to the abdomen or back. (2) An episode of hypotension in the elderly patient, no matter how transient, must be taken seriously and should suggest RAAA as a possibility, because this may be the herald sign that retroperitoneal hemorrhage is occurring. (3) Thorough examination of the abdomen for a pulsarile mass must be performed in all elderly patients. (4) Men over the age of 50 years admitted with new onset of lower back pain and no definite precipitating event should be evaluated for an AAA, at least with a thorough abdominal examination by an experienced examiner. Most misdiagnosed patients underwent diagnostic radiography: plain abdominal films were obtained of nearly two thirds, abdominal CT scans of 25%, and sonograms or arteriograms of a few patients each. Plain abdominal films are unlikely to provide a definite diagnosis of RAAA, but Loughran, 16 in a retrospective review of 31 patients with RAAA, reports that evidence of the presence of the aneurysm or retroperitoneal hemmorhage was present on supine abdominal films in nearly 90%. The most frequently noted signs were calcification of the aneurysm in 65%, complete loss of one or both psoas outlines in 75%, and complete loss of one or both renal outlines in 78%. Other diagnostic modalities frequently obtained are abdominal CT scans, sonography, and arteriography. In patients who are in shock, none of these studies is indicated, because they delay rapid surgical
Misdiagnosis of ruptured abdmninal aortic aneurysms 21
treatment. In patients who are hemodynamically stable, the diagnostic procedure of choice would be the one most rapidly available, either sonography or CT scanning. Both are sensitive in identifying the presence of an AAA, lz,ls although CT scanning is more likely than ultrasonography to provide informarion on the presence of rupture and the extent of retroperitoneal hematoma. Furthermore, CT scanning may be more readily available in hospitals at night or on weekends. The sensitivity of CT scanning in identifying rupture has been reported as 90% to 94% and the specificity as 92% to 1 0 0 % . 1719 However, sonography has the advantage of being portable. It could be used in patients, even those with moderate hypotension, in whom the diagnosis of RAAA is in doubt. These patients could be taken directly to the operating room, where sonography could be performed while preparations for laparotomy are made. Surgery would proceed on diagnosis, or on an emergency basis if the patient's condition were to deteriorate. The impact of misdiagnosis on the mortality rate of RAAA was difficult to determine. The overall mortality rate was slightly lower than, but not significantly different from, those correctly diagnosed. However, apparently two subgroups existed within the population. These were divided on the basis of rime of diagnostic delay. In those patients with a diagnostic delay greater than 8 hours the incidence of shock was only 20%, with only three of 20 sustaining free ruptures and a relatively low mortality rate. This group's mortality rate is similar to that reported for symptomatic unruptured AAAs undergoing emergency operation. This was reported at 8% to 26%, and the increased mortality rate compared with elective cases was blamed on the lack of adequate resuscitation and cardiovascular preparation for surgery, leading to an increased incidence of myocardial infarction. 20"22 Certainly, in patients with KAAA, delay in operation for resuscitation and maximizing cardiac function is not an option. Therefore it may be difficult to improve the outcome in this group with our present methods of management. Patients whose delay in diagnosis is short (less than 8 hours) have a significantly greater incidence of shock, with nearly half having free ruptures and a mortality rate of 62%. This is similar to the entire group of correctly diagnosed patients, who had an incidence of free rupture of 43% and a mortality rate of 58%. The apparent better survival in the group of patients with prolonged delay in diagnosis is misleading. The true denominator of the number of
22
Journal of VASCULAR SURGERY
A/larston et al.
patients with RAAA for that period of time is unknown, and deaths incorrectly attributed to other causes would alter the true mortality rate of this group. Furthermore, this subgroup of patients is not the same as those who were unstable and had emergency surgery or died earlier. The patients who survived to have prolonged duration ofmisdiagnosis were preselected by their ability to tolerate some degree of blood loss. The patients who had greater blood loss or could not compensate, or both, would be eliminated by early death and thereby artificially improve survival with delayed diagnosis. The urgency to operate immediately when a rupture is diagnosed is illustrated by the fact that 31% of the patients with delays greater than 8 hours who were initially stable later developed shock. Two of the five had free ruptures, and three died. Whether these patients initially had nonruptured, symptomatic aneurysms that later ruptured or small initial ruptures that progressed to hemodynarnic significance is not known.
In conclusion, in this large series of RAAAs we have found that misdiagnosis of RAAA is a frequent occurrence leading to a delay in diagnosis. These patients commonly were admitted with either back pain, abdominal pain, or shock, but rarely with all three. The absence ofa pulsatile abdominal mass was found to be the most likely cause of an incorrect diagnosis. The impact of misdiagnosis on mortality rates was difficult to determine. The overall mortality rate of patients who were misdiagnosed was not significantly different from those correctly diagnosed. However, those with long delay times had a lower mortality rate, probably because they constituted a group preselected for less severe ruptures in patients who could survive through the prolonged delay to operative intervention. The diagnosis of RAAA should be considered in all elderly patients with symptoms related to the abdomen or back and should be rapidly investigated and treated if RAAA is a likely possibility. REFERENCES
1. Crawford ES, Saleh SA, Babb JR III, Glaeser DH, Vaccaro PS, Silvers A. Infrarenal abdominal aortic aneurysm: factors influencing survival after operation performed over a 25-year period. Ann Surg 1981;193:699-709. 2. Meyer AA, Ahlquist RE Jr, Trunkey DD. Mortality from ruptured abdominal aortic aneurysms: a comparison of two series. Am J Surg 1986;152:27-33. 3. Bodily KC, Buttorff JD. Ruptured abdominal aortic aneurysm: the Tacoma experience. Am J Surg 1985;149: 580-2.
4. Hoffman M, Avellonc JC, Plecha FR, et al. Operation fbr ruptured abdominal aortic aneurysms: a community-wide experience. Surgery 1982;91:597-602. 5. Wakefield TW, Whitehouse WM Jr, Wu SC~ et al. Abdominal aortic aneurysm rupture: statistical analysis of factors affecting outcome of surgical treatment. Surgery 1982;91:586-96. 6. Lawrie GM, Morris GC Jr, Crawford ES, et al. Improved results of operation for ruptured abdominal aortic aneurysms. Surgery 1979;85:453-8. 7. Lawler M Jr. Aggressive treatment of ruptured abdominal aortic aneurysm in a community hospital. Surgery 1984;95: 38-44. 8. Johnson G Jr, Avery A, McDougal EG, Burnham SJ, Keagy BA. Aneurysms of the abdominal aorta: incidence in blacks and whites in North Carolina. Arch Surg 1985;120:1138-40. 9. Darling RC. Ruptured arteriosclerotic abdominal aortic aneurysms: a pathologic and clinical study. Am J Surg 1970;119:397-401. 10. McGregor JC. Unoperated ruptured abdominal aortic aneurysms: a retrospective clinicopathological study over a 10-year period. Br J Surg 1976;63:113-6. 11. Sterpetti AV, Blair EA, Schultz RD. Sealed rupture of abdominal aortic aneurysms. J VASC SURG 1990;11:430-5. 12. Louras JC, Welch JP. Masking of ruptured abdominal aortic aneurysm by incarcerated inguinal hernia. Arch Surg 1984; 119:331-2. 13. Grabowski EW, Pilcher DB. Ruptured abdominal aortic aneurysm manifesting as symptomatic inguinal hernia. Am Surg 1981;47:311-2. 14. Merchant RF, Cafferta HT, DePalma RG. Ruptured abdominal aortic aneurysm seen initiallyas acute femoral neuropathy. Arch Surg 1982;117:811-3. 15. Chandler JJ. The Einstein sign: the clinical picture of acute cholecystitis caused by ruptured abdominal aortic aneurysm. N Engl J Med 1984;310:1538. 16. Loughran CF. A review of the plain abdominal radiograph in acute rupture of abdominal aortic aneurysms. Clin Radiol 1986;37:383-7. 17. Zarnke MD, Gould HR, Goldman MH. Computed tomography in the evaluation of the patient with symptomatic abdominal aortic aneurysm. Surgery 1988;103:638-42. 18. Kvilekval KH, Best IM, Mason RA, et al. The value of computed tomography in the management of symptomatic abdominal aortic aneurysms. J VAse: SURe 1990;12: 28-33. 19. Weinbaum FI, Dubner S, Turner JW, Pardes JG. The accuracy of computed tomography in the diagnosis of retroperitoneal blood in the presence of abdominal aortic aneurysm. J VAsc SURG 1987;6:11-6. 20. Johnson G Jr, McDevitt NB, Proctor HJ, Mandel SR, Peacock JB. Emergent or elective operation for symptomatic abdominal aortic aneurysm. Arch Surg 1980;115:51-3. 21. Sullivan CA, Rohrer MJ, Cutler BS. Clinical management of the symptomatic but unruptured abdominal aortic aneurysm. J VASC SURG 1990;11:799-803. 22. Payne DF Jr, Rosenthal D, Lamis PA, Stanton PE Jr. Infrarenal aortic aneurysms: asyrnptomatic versus symptomatic. Am Surg 1985;51:94-6. Submitted July 26, 1991; accepted Oct. 11, 1991.
Spokane, Wash. Ruptured abdominal aortic aneurysm is a surgical emergency with a high mortality rate even when diagnosed and repaired immediately. We retrospectivelyreviewed 152 cases of ruptured abdominal aneurysms to identify the incidence of misdiagnosis leading to a delay in treatment, the most frequent misdiagnoses, and the outcome in this group of patients. Forty-six (30%) were initially misdiagnosed. The most common misdiagnoses were renal colic, diverticulitis, and gastrointestinal hemorrhage. The most common initial physical findings in misdiagnosed patients were abdominal pain (70%), shock (57%), and back pain (50%). A pulsatile abdominal mass was found in only 26% of misdiagnosed patients versus 72% of patients correctly diagnosed (p < 0.005). Misdiagnosed ruptured abdominal aneurysm had a 44% mortality rate, which was not significantly different from patients correctly diagnosed (58%,p = 0.34). The lack of difference in mortality rates is most likely due to preselection of those miscliagnosed patients who were able to withstand the delay in diagnosis and survive to surgical treatment. The 30% incidence of misdiagnosis in this series suggests that it is frequently a difficult diagnosis to make and must be considered in elderly patients, especially men, who are admitted with abdominal pain and/or back pain. (J VAse SURG 1992;16:17-22.) Rupture of abdominal aortic aneurysm (RAAA) is a catastrophic event that was a uniformly fatal complication of AAA until the 1950s, when emergent repair of RAAA first became possible. Results improved significantly in the next 15 years as a result of improvements in surgical, anesthetic, and postoperative management. However, the mortality rate remains quite high, reported at 1596 to 7096 in various series, and has not improved in the last 2 decades.l"z Early diagnosis is believed to be important in improving the outcome of RAAA, Several investigators examining the causes of death have identified an increased mortality rate for persons who are incorrectly diagnosed on admission, s'4,6 The frequency of misdiagnosis of RAAA is not well known nor are the medical diagnoses that are commonly confused with RAAA. In this article reports of RAAAs from three institutions were reviewed to
From the Departments of Surgery, the Universityof North Carolina Hospitals, Universityof California-San Francisco, and Spokane. Presentedat the North CarolinaChapterof the AmericanCollege of Surgeons,Asheville,N.C., July 14, 1991. Reprint requests: AnthonyA. Meyer,MD, PhD, Universityof North CarolinaDepartmentof Surgery,164 Burnett-Womack, Campus Box 7210, ChapelHill, NC 27599-7210. 24/1/34344
identify the incidence of misdiagnosis and characteristics of patients whose RAAAs are misdiagnosed. METHODS Patients. One hundred fifty-two patients from three institutions who sustained RAAAs between 1974 and 1989 were diagnosed either in a referring emergency department or at the regional referral center. Cases in which the rupture occurred during elective AAA repair were excluded, as were patients with no evidence of retroperitoneal or free bleeding at operation. The three institutions involved were the University of North Carolina Hospitals, a vascular referral center (58 cases); San Francisco General Hospital, a municipal teaching hospital (47 cases); and a community hospital practice in Spokane, Wash. (47 cases). Record review. Emergency depamnent records of both the initial and referral centers were reviewed for each patient to identify initial symptoms, hemodynamic stability, initial diagnosis, time until diagnosis of RAAA, and diagnostic studies performed. Misdiagnosis of RAAA was defined as a delay in diagnosis of at least 6 hours or an initial primary diagnosis other than RAAA. Two of the following three symptoms were required to diagnose shock: blood pressure less than 70 mm Hg, pulse greater
17
Journalof VASCULAR SURGERY
18 Alarston et al.
Table I. Findings on initial examination Diagnosis
No. of patients
Shock
Bkpn
Abdpn
Mass
Avg delay
Mortality rate
Renal colic Diverticulitis GI hemorrhage Acute MI Back pain MVA Sepsis Other GI Other/no Dx Total
11 6 6 4 4 3 3 3 6 46
6 3 3 2 2 3 3 3 1 26
9 2 0 2 4 2 0 2 2 23
9 6 5 2 1 2 1 3 3 32
1 1 2 2 0 2 2 0 2 12
15 hr 79 hr 17 hr 13 hr 18 hr 1.5 hr 26 hr 4 hr 18 hr 24.4
5 (45%) 1 (17%) 3 (50%) 2 (50%) 2 (50%) 3 (100%) 2 (67%) 1 (33%) 1 (17%) 20 (44%)
Bkpn, Back pain; Abdpn, abdominal pain; Mass, pulsatile mass on first examination; Avg delay, average delay in diagnosis; A4/, myocardial infarction; MVA, motor vehicle accident; G/, gastrointestinal; Dx, diagnosis.
than 125 beats per minute, or poor peripheral perfusion. For those patients identified with misdiagnosis of RAAA, hospital records were reviewed to determine operative characteristics and outcome. Mortality was defined as death before discharge from the hospital or within 30 days of surgery. Statistics. Differences between groups were tested by chi-square analysis by means of a microcomputer, with a p value of _ 8 hr (n = 20)
p value
22 (84.6%) 11 (42.3%) 16 (62%)
4 (20%) 3 (15%) 4 (20%)
0.015 0.13 0.07
series, and is reported to be increased by delay in operative treatment. 2.s We examined 46 cases of misdiagnosis to identify factors leading to misdiagnosis and to evaluate the effect of delay in treatment on outcome. Patients admitted with RAAA are typically elderly men, with 81% being 64 years of age or older. The symptoms most commonly found in this review of 152 cases were abdominal and back pain, at 76% and 54%, respectively. Both were present in only 42%, and the classic diagnostic triad of abdominal pain, back pain, and pulsatile mass was found in only 26% of patients. In those patients who were misdiagnosed, the classic diagnostic triad was found in only 9% of cases. Misdiagnosis was more frequent in those patients admitted initially to a referring emergency depat-ti~tent than in those admitted initially to a tertiary care center. When the frequency of symptoms was compared in patients correctly diagnosed versus those misdiagnosed, no significant difference was found in the incidence of abdominal or back pain. The incidence of shock was slightly higher in those correctly diagnosed. A previously known AAA was infrequent in both groups. However, the detection ofa pulsatile mass was much more frequent in those correctly diagnosed, with only one fourth of the misdiagnosed patients having a pulsatile mass on initial examination. Another 25% of misdiagnosed patients were found to have a mass on a later examination, usually performed by a surgical consultant. A pulsatile mass may not be identifiable in many patients. Obesity often obscures this finding, and many patients may
have enough bleeding in the retroperitoneum to cause the whole abdomen to have a distended, doughy feel. Furthermore, profoundly hypotensive patients may not generate a systolic pressure high enough to palpate a pulsatile mass. Only 14% of 88 patients with a pulsatile mass on initial physical examination were misdiagnosed, whereas 53% of the 64 patients with no pulsatile mass were misdiagnosed (p < 0.0003). In patients whose aneurysm did rupture, the presence ofa pulsatile mass was the most important finding aiding correct and rapid diagnosis, emphasizing the need for a prompt abdominal examination in all elderly male patients, by a surgical consultant if possible. The erroneous diagnoses most frequently made were those associated with abdominal pain, shock, and back pain, the symptoms commonly seen in this series. Diagnoses of intraabdominal processes, primarily sigmoid diverticulitis and gastrointestinal hemorrhage were made in nearly one third of cases. The incidence of gastrointestinal hemorrhage as a diagnostic error is partially due to the association of aortoenteric fistula with rupture in three patients in this study. Patients misdiagnosed with diverticulitis generally had long delays in diagnosis, usually until an abdominal CT scan was obtained when their symptoms did not improve on conservative therapy. These patients did well as a group, probably because of their hemodynamic stability and their improved volume status after several days of intravenous hydration before correct diagnosis of R A A . However, patients with RAAA unresponsive to such limited resuscitation may die with this incorrect diagnosis, and the true cause of death may never be known. Renal colic was also a frequent misdiagnosis, with most patients having severe back pain and two thirds having both back and abdominal pain. More than 50% had a hypotensive episode during the initial evaluation, but most of these were transient events, diagnosed as a vagal episode or a result of developing sepsis. In some cases the diagnosis was not made until
Volume 16 Number 1 July 1992
a second, more severe hypotensive episode occurred, suggesting that a more acute process was developing. Acute myocardial infarction, lower back pain, sepsis, and blunt motor vehicle trauma were the primary diagnoses in several cases each. Patients with RAAA may have myocardial ischemia, with characteristic changes on an electrocardiogram, probably caused by hypotension in the patient with diffuse atherosclerotic disease, which involved the coronary arteries as well as the aorta. Three of the four patients misdiagnosed with lower back pain had no abdominal examination documented on their initial examination record. In patients admitted with presumed blunt trauma after a motor vehicle accident, the degree of vehicular damage was mild in comparison with their symptoms. All three were in shock, but none had severe long bone fractures, significant chest injuries, or other evidence of major injury. General observations made from reviewing these patients' presentations are the following: (1) In elderly patients, especially white men, the possibility of symptomatic or RAAA must be considered in all patients who are admitted with any symptoms related to the abdomen or back. (2) An episode of hypotension in the elderly patient, no matter how transient, must be taken seriously and should suggest RAAA as a possibility, because this may be the herald sign that retroperitoneal hemorrhage is occurring. (3) Thorough examination of the abdomen for a pulsarile mass must be performed in all elderly patients. (4) Men over the age of 50 years admitted with new onset of lower back pain and no definite precipitating event should be evaluated for an AAA, at least with a thorough abdominal examination by an experienced examiner. Most misdiagnosed patients underwent diagnostic radiography: plain abdominal films were obtained of nearly two thirds, abdominal CT scans of 25%, and sonograms or arteriograms of a few patients each. Plain abdominal films are unlikely to provide a definite diagnosis of RAAA, but Loughran, 16 in a retrospective review of 31 patients with RAAA, reports that evidence of the presence of the aneurysm or retroperitoneal hemmorhage was present on supine abdominal films in nearly 90%. The most frequently noted signs were calcification of the aneurysm in 65%, complete loss of one or both psoas outlines in 75%, and complete loss of one or both renal outlines in 78%. Other diagnostic modalities frequently obtained are abdominal CT scans, sonography, and arteriography. In patients who are in shock, none of these studies is indicated, because they delay rapid surgical
Misdiagnosis of ruptured abdmninal aortic aneurysms 21
treatment. In patients who are hemodynamically stable, the diagnostic procedure of choice would be the one most rapidly available, either sonography or CT scanning. Both are sensitive in identifying the presence of an AAA, lz,ls although CT scanning is more likely than ultrasonography to provide informarion on the presence of rupture and the extent of retroperitoneal hematoma. Furthermore, CT scanning may be more readily available in hospitals at night or on weekends. The sensitivity of CT scanning in identifying rupture has been reported as 90% to 94% and the specificity as 92% to 1 0 0 % . 1719 However, sonography has the advantage of being portable. It could be used in patients, even those with moderate hypotension, in whom the diagnosis of RAAA is in doubt. These patients could be taken directly to the operating room, where sonography could be performed while preparations for laparotomy are made. Surgery would proceed on diagnosis, or on an emergency basis if the patient's condition were to deteriorate. The impact of misdiagnosis on the mortality rate of RAAA was difficult to determine. The overall mortality rate was slightly lower than, but not significantly different from, those correctly diagnosed. However, apparently two subgroups existed within the population. These were divided on the basis of rime of diagnostic delay. In those patients with a diagnostic delay greater than 8 hours the incidence of shock was only 20%, with only three of 20 sustaining free ruptures and a relatively low mortality rate. This group's mortality rate is similar to that reported for symptomatic unruptured AAAs undergoing emergency operation. This was reported at 8% to 26%, and the increased mortality rate compared with elective cases was blamed on the lack of adequate resuscitation and cardiovascular preparation for surgery, leading to an increased incidence of myocardial infarction. 20"22 Certainly, in patients with KAAA, delay in operation for resuscitation and maximizing cardiac function is not an option. Therefore it may be difficult to improve the outcome in this group with our present methods of management. Patients whose delay in diagnosis is short (less than 8 hours) have a significantly greater incidence of shock, with nearly half having free ruptures and a mortality rate of 62%. This is similar to the entire group of correctly diagnosed patients, who had an incidence of free rupture of 43% and a mortality rate of 58%. The apparent better survival in the group of patients with prolonged delay in diagnosis is misleading. The true denominator of the number of
22
Journal of VASCULAR SURGERY
A/larston et al.
patients with RAAA for that period of time is unknown, and deaths incorrectly attributed to other causes would alter the true mortality rate of this group. Furthermore, this subgroup of patients is not the same as those who were unstable and had emergency surgery or died earlier. The patients who survived to have prolonged duration ofmisdiagnosis were preselected by their ability to tolerate some degree of blood loss. The patients who had greater blood loss or could not compensate, or both, would be eliminated by early death and thereby artificially improve survival with delayed diagnosis. The urgency to operate immediately when a rupture is diagnosed is illustrated by the fact that 31% of the patients with delays greater than 8 hours who were initially stable later developed shock. Two of the five had free ruptures, and three died. Whether these patients initially had nonruptured, symptomatic aneurysms that later ruptured or small initial ruptures that progressed to hemodynarnic significance is not known.
In conclusion, in this large series of RAAAs we have found that misdiagnosis of RAAA is a frequent occurrence leading to a delay in diagnosis. These patients commonly were admitted with either back pain, abdominal pain, or shock, but rarely with all three. The absence ofa pulsatile abdominal mass was found to be the most likely cause of an incorrect diagnosis. The impact of misdiagnosis on mortality rates was difficult to determine. The overall mortality rate of patients who were misdiagnosed was not significantly different from those correctly diagnosed. However, those with long delay times had a lower mortality rate, probably because they constituted a group preselected for less severe ruptures in patients who could survive through the prolonged delay to operative intervention. The diagnosis of RAAA should be considered in all elderly patients with symptoms related to the abdomen or back and should be rapidly investigated and treated if RAAA is a likely possibility. REFERENCES
1. Crawford ES, Saleh SA, Babb JR III, Glaeser DH, Vaccaro PS, Silvers A. Infrarenal abdominal aortic aneurysm: factors influencing survival after operation performed over a 25-year period. Ann Surg 1981;193:699-709. 2. Meyer AA, Ahlquist RE Jr, Trunkey DD. Mortality from ruptured abdominal aortic aneurysms: a comparison of two series. Am J Surg 1986;152:27-33. 3. Bodily KC, Buttorff JD. Ruptured abdominal aortic aneurysm: the Tacoma experience. Am J Surg 1985;149: 580-2.
4. Hoffman M, Avellonc JC, Plecha FR, et al. Operation fbr ruptured abdominal aortic aneurysms: a community-wide experience. Surgery 1982;91:597-602. 5. Wakefield TW, Whitehouse WM Jr, Wu SC~ et al. Abdominal aortic aneurysm rupture: statistical analysis of factors affecting outcome of surgical treatment. Surgery 1982;91:586-96. 6. Lawrie GM, Morris GC Jr, Crawford ES, et al. Improved results of operation for ruptured abdominal aortic aneurysms. Surgery 1979;85:453-8. 7. Lawler M Jr. Aggressive treatment of ruptured abdominal aortic aneurysm in a community hospital. Surgery 1984;95: 38-44. 8. Johnson G Jr, Avery A, McDougal EG, Burnham SJ, Keagy BA. Aneurysms of the abdominal aorta: incidence in blacks and whites in North Carolina. Arch Surg 1985;120:1138-40. 9. Darling RC. Ruptured arteriosclerotic abdominal aortic aneurysms: a pathologic and clinical study. Am J Surg 1970;119:397-401. 10. McGregor JC. Unoperated ruptured abdominal aortic aneurysms: a retrospective clinicopathological study over a 10-year period. Br J Surg 1976;63:113-6. 11. Sterpetti AV, Blair EA, Schultz RD. Sealed rupture of abdominal aortic aneurysms. J VASC SURG 1990;11:430-5. 12. Louras JC, Welch JP. Masking of ruptured abdominal aortic aneurysm by incarcerated inguinal hernia. Arch Surg 1984; 119:331-2. 13. Grabowski EW, Pilcher DB. Ruptured abdominal aortic aneurysm manifesting as symptomatic inguinal hernia. Am Surg 1981;47:311-2. 14. Merchant RF, Cafferta HT, DePalma RG. Ruptured abdominal aortic aneurysm seen initiallyas acute femoral neuropathy. Arch Surg 1982;117:811-3. 15. Chandler JJ. The Einstein sign: the clinical picture of acute cholecystitis caused by ruptured abdominal aortic aneurysm. N Engl J Med 1984;310:1538. 16. Loughran CF. A review of the plain abdominal radiograph in acute rupture of abdominal aortic aneurysms. Clin Radiol 1986;37:383-7. 17. Zarnke MD, Gould HR, Goldman MH. Computed tomography in the evaluation of the patient with symptomatic abdominal aortic aneurysm. Surgery 1988;103:638-42. 18. Kvilekval KH, Best IM, Mason RA, et al. The value of computed tomography in the management of symptomatic abdominal aortic aneurysms. J VAse: SURe 1990;12: 28-33. 19. Weinbaum FI, Dubner S, Turner JW, Pardes JG. The accuracy of computed tomography in the diagnosis of retroperitoneal blood in the presence of abdominal aortic aneurysm. J VAsc SURG 1987;6:11-6. 20. Johnson G Jr, McDevitt NB, Proctor HJ, Mandel SR, Peacock JB. Emergent or elective operation for symptomatic abdominal aortic aneurysm. Arch Surg 1980;115:51-3. 21. Sullivan CA, Rohrer MJ, Cutler BS. Clinical management of the symptomatic but unruptured abdominal aortic aneurysm. J VASC SURG 1990;11:799-803. 22. Payne DF Jr, Rosenthal D, Lamis PA, Stanton PE Jr. Infrarenal aortic aneurysms: asyrnptomatic versus symptomatic. Am Surg 1985;51:94-6. Submitted July 26, 1991; accepted Oct. 11, 1991.
