ORIGINAL CONTRIBUTION ectopic pregnancy
Emergency Department Diagnosis of Ectopic Pregnancy Study objectives: To assess the accuracy of the history and physical examination as compared to the addition of serum progesterone screening for ectopic pregnancy in women presenting to the emergency department. Design: Prospective, consecutive case series, N~2,157. Setting: ED of the Regional Medical Center at Memphis, a publicly subsidized, 450-bed acute care hospital staffed by residents and faculty of the University of Tennessee, Memphis. Type of participants: All ED patients with a positive urine pregnancy test treated between January 1 and December 31, 1988. Interventions: Screening history, physical examination, and serum progesterone (P) and quantitative human chorionic gonadotropin (hCG) titer. Measurements: All discharged patients were given follow-up appointments within two weeks; those found to have a P < 25 ng/mL were called to return for repeat hCG and transvaginal ultrasound. Main results: One hundred sixty-one of 2,157 patients (7.5%) with a positive urine pregnancy test were found to have an ectopic pregnancy, All but five had a P of < 25 ng/mL (sensitivity, 97%); four of these were admitted for immediate surgery because of symptoms. Overall, the ED physician detected 89 of 161 ectopics (55.3%) on initial presentation, 53 (60%) of which were ruptured at the t i m e of surgery. Seventy-two patients (44.7%) who were discharged but later found to have an ectopic pregnancy had benign clinical presentations, including 41 with vaginal bleeding. There were no statistically significant differences in the presenting symptoms of patients with unruptured ectopics compared with normal intrauterine pregnancies. All but one of the 72 discharged patients were noted the following day to have a progesterone of < 25 ng/mL and contacted to return. Eight of these were found to have a ruptured ectopic at the time of surgery, Only 91 of 161 patients (56.5%) with ectopic pregnancy acknowledged one or more clinical risk factors on follow-up questioning. Conclusion: The standard history and physical examination, including those performed by gynecologic specialists, are insufficiently sensitive for early detection of unruptured ectopic pregnancy, EDs with a high incidence of ectopic pregnancy should strongly consider implementation of a universal progesterone screening program to decrease unnecessary patient morbidity and the risk of mortality from undiagnosed ectopic pregnancy. [Stovail TG, Kellerman AL, Ling FW, Buster JE: Emergency department diagnosis of ectopic pregnancy. Ann Emerg Med October 1990;19:1098-
Thomas G Stovall, MD* Arthur L Kellerman, MD, MPH:~ Frank W Ling, MD* John E Buster, MDt Memphis, Tennessee From the Department of Obstetrics and Gynecology, Divisions of Gynecology* and Reproductive Endocrinology,t and the Department of Medicine, Division of Emergency Medicine,¢ University of Tennessee, Memphis. Received for publication October 2, 1989. Revision received February 5, 1990. Accepted for publication February 26, 1990. Presented at the Society for Academic Emergency Medicine Annual Meeting in San Diego, May 1989. Address for reprints: Thomas G Stovall, MD, Department of Obstetrics and Gynecology, 853 Jefferson Avenue, Room E-102, Memphis, Tennessee 38163.
1103.]
INTRODUCTION The incidence of ectopic pregnancy in the United States has increased fourfold since 1970.1 Current rates of ectopic pregnancy are estimated to range from one in 200 to one in 45 live births, depending on the population studied. 2 More than 80,000 w o m e n are hospitalized with a diagnosis of ectopic pregnancy each year. 3 Black w o m e n are disproportionately affected; of the 36 deaths recorded in 1986, 19 were black women. 4 Early diagnosis of ectopic pregnancy can substantially reduce a w o m a n ' s risk of morbidity and mortality and increase her chances of childbearing in the future. The use of outpatient laparoscopic surgical treatment has reduced the cost of treatment, and more recently the development of a medical approach to management has permitted nonsurgical outpatient treatment, s 19:10 October 1990
Annals of Emergency Medicine
1098/49
ECTOPIC PREGNANCY Stovall et al
History Abdomina VPelvic Pain
100%
Amenorrhea
One or More Husk f ,1(:1ors t'resenl
%
I "NS
6o 7 %
Physical Exam
V;l(Iinal Bh)edlng 902
%
[]
Inlraulerm,9 Prognan('y N-34!}
[]
t/nruptured I ('topK: N - I O 0
•
ftuplured | ct(~pl(: N-61
Adnc, xal I endefness
100%
A(fiwxdl Mass
492
%
('ewR'al M o l ~ n 1 ('rl(Jur rli~,g.'g
1
lOO%
o
20
FIGURE 1. Frequency of history and physical examination findings in pat i e n t s w i t h r u p t u r e d e c t o p i c pregn a n c y versus an unruptured ectopic pregnancy versus a norma] intrauterine pregnancy. Traditionally, physicians have relied on history and physical examination to screen for possible cases of ectopic pregnancy. When the diagnosis is unclear, use of serial quantitative serum chorionic gonadotropin levels (hCG) is advocated for those suspected of ectopic pregnancy. This test requires two blood draws separated by a period of 48 hours. 6 Vaginal ultrasound can detect or exclude the presence of an intrauterine sac as early as five weeks after the last m e n s t r u a l period, but the test is highly operator dependent and in m a n y h o s p i t a l s is u n a v a i l a b l e at night and weekends.7, 8 In addition, ultrasound findings often must be correlated with the hCG titer. In 1978, Radwanska et al noted an association between low levels of serum progesterone and abnormal pregnancy. 9 This observation was subsequently confirmed by others in retrospective studies.m, 1] In contrast to serial hCG titers, serum progesterone is not gestational age dependent. Be50/1099
40
GO
80
lOO
cause only a single level is needed, the requirement for a specified time between samples in order to observe a trend in values is eliminated. Following promising results with a progesterone screening program, 12 we sought to compare the value of the clinical history and physical examination to that derived from progesterone screening in a large consecutive series of patients presenting to an emergency d e p a r t m e n t with a high patient incidence of ectopic pregnancy. METHODS This study was conducted in the ED of the Regional Medical Center at M e m p h i s (MED) a 450-bed, acute care h o s p i t a l o w n e d by S h e l b y County Government, operated by a not-for-profit health care corporation and staffed by residents and faculty of the U n i v e r s i t y of T e n n e s s e e , Memphis. As a publicly subsidized facility, the MED provides care to all Shelby County citizens in need, regardless of their ability to pay. The MED ED is the county's busiest and treats more than 67,000 patients annually. Stable "walk-in" patients are triaged to the ED's Acute Ambulatory Care Clinic, where they are evaluated by PGY-1 residents from the DeAnnals of Emergency Medicine
120
partments of Medicine, Surgery, or Obstetrics and Gynecology. Attending supervision is provided at all times. In-house gynecology consultation is also available on request. All reproductive-aged female patients whose last menstrual period was at least 25 days prior to their visit, w h o had a b n o r m a l uterine bleeding or pelvic pain, or when a diagnostic test would be contraindicated in pregnancy had urine collected for hCG testing (Tandem Icon II hCG I m m u n o e n z y m e t r i c Assay, Hybritech Inc, San Diego, California). T h o s e p a t i e n t s p r e s e n t i n g to the MED ED between January 1 and December 31, 1988, and noted to have a positive urine hCG were eligible for inclusion in the study. Pelvic examinations were performed when indicated for gynecologic symptoms (eg, abdominal/pelvic pain, vaginal bleeding or spotting, suspected vaginitis or suspected sexually transmitted disease. )
Serum was also obtained for quantitative analysis of h C G and progesterone levels. If the patient's history and/or physical e x a m i n a t i o n was suggestive of ectopic pregnancy, g y n e c o l o g y c o n s u l t a t i o n was requested and the patient was admitted or discharged at the discretion of this consulting service. 19:10 October 1990
TABLE 1. History and physical examination findings in patients diagnosed with an ectopic pregnancy at their initial ED visit compared with those diagnosed on follow-up
Initial Diagnosis N = 89 (%)
Follow-up Diagnosis N = 72 (%)
P
Vaginal bleeding Amenorrhea
75 (84.3) 48 (53.9)
41 (56.9) 52 (72.2)
NS NS
Pain Cervical motion tenderness Adnexal tenderness Adnexal mass
81 80 82 34
10 6 7 0
History and Physical Examination
(91.0) (89.9) (92.1) (38.2)
(13.9) (8.3) (9.7) (0.0)
< < <
Emergency Department Diagnosis of Ectopic Pregnancy Study objectives: To assess the accuracy of the history and physical examination as compared to the addition of serum progesterone screening for ectopic pregnancy in women presenting to the emergency department. Design: Prospective, consecutive case series, N~2,157. Setting: ED of the Regional Medical Center at Memphis, a publicly subsidized, 450-bed acute care hospital staffed by residents and faculty of the University of Tennessee, Memphis. Type of participants: All ED patients with a positive urine pregnancy test treated between January 1 and December 31, 1988. Interventions: Screening history, physical examination, and serum progesterone (P) and quantitative human chorionic gonadotropin (hCG) titer. Measurements: All discharged patients were given follow-up appointments within two weeks; those found to have a P < 25 ng/mL were called to return for repeat hCG and transvaginal ultrasound. Main results: One hundred sixty-one of 2,157 patients (7.5%) with a positive urine pregnancy test were found to have an ectopic pregnancy, All but five had a P of < 25 ng/mL (sensitivity, 97%); four of these were admitted for immediate surgery because of symptoms. Overall, the ED physician detected 89 of 161 ectopics (55.3%) on initial presentation, 53 (60%) of which were ruptured at the t i m e of surgery. Seventy-two patients (44.7%) who were discharged but later found to have an ectopic pregnancy had benign clinical presentations, including 41 with vaginal bleeding. There were no statistically significant differences in the presenting symptoms of patients with unruptured ectopics compared with normal intrauterine pregnancies. All but one of the 72 discharged patients were noted the following day to have a progesterone of < 25 ng/mL and contacted to return. Eight of these were found to have a ruptured ectopic at the time of surgery, Only 91 of 161 patients (56.5%) with ectopic pregnancy acknowledged one or more clinical risk factors on follow-up questioning. Conclusion: The standard history and physical examination, including those performed by gynecologic specialists, are insufficiently sensitive for early detection of unruptured ectopic pregnancy, EDs with a high incidence of ectopic pregnancy should strongly consider implementation of a universal progesterone screening program to decrease unnecessary patient morbidity and the risk of mortality from undiagnosed ectopic pregnancy. [Stovail TG, Kellerman AL, Ling FW, Buster JE: Emergency department diagnosis of ectopic pregnancy. Ann Emerg Med October 1990;19:1098-
Thomas G Stovall, MD* Arthur L Kellerman, MD, MPH:~ Frank W Ling, MD* John E Buster, MDt Memphis, Tennessee From the Department of Obstetrics and Gynecology, Divisions of Gynecology* and Reproductive Endocrinology,t and the Department of Medicine, Division of Emergency Medicine,¢ University of Tennessee, Memphis. Received for publication October 2, 1989. Revision received February 5, 1990. Accepted for publication February 26, 1990. Presented at the Society for Academic Emergency Medicine Annual Meeting in San Diego, May 1989. Address for reprints: Thomas G Stovall, MD, Department of Obstetrics and Gynecology, 853 Jefferson Avenue, Room E-102, Memphis, Tennessee 38163.
1103.]
INTRODUCTION The incidence of ectopic pregnancy in the United States has increased fourfold since 1970.1 Current rates of ectopic pregnancy are estimated to range from one in 200 to one in 45 live births, depending on the population studied. 2 More than 80,000 w o m e n are hospitalized with a diagnosis of ectopic pregnancy each year. 3 Black w o m e n are disproportionately affected; of the 36 deaths recorded in 1986, 19 were black women. 4 Early diagnosis of ectopic pregnancy can substantially reduce a w o m a n ' s risk of morbidity and mortality and increase her chances of childbearing in the future. The use of outpatient laparoscopic surgical treatment has reduced the cost of treatment, and more recently the development of a medical approach to management has permitted nonsurgical outpatient treatment, s 19:10 October 1990
Annals of Emergency Medicine
1098/49
ECTOPIC PREGNANCY Stovall et al
History Abdomina VPelvic Pain
100%
Amenorrhea
One or More Husk f ,1(:1ors t'resenl
%
I "NS
6o 7 %
Physical Exam
V;l(Iinal Bh)edlng 902
%
[]
Inlraulerm,9 Prognan('y N-34!}
[]
t/nruptured I ('topK: N - I O 0
•
ftuplured | ct(~pl(: N-61
Adnc, xal I endefness
100%
A(fiwxdl Mass
492
%
('ewR'al M o l ~ n 1 ('rl(Jur rli~,g.'g
1
lOO%
o
20
FIGURE 1. Frequency of history and physical examination findings in pat i e n t s w i t h r u p t u r e d e c t o p i c pregn a n c y versus an unruptured ectopic pregnancy versus a norma] intrauterine pregnancy. Traditionally, physicians have relied on history and physical examination to screen for possible cases of ectopic pregnancy. When the diagnosis is unclear, use of serial quantitative serum chorionic gonadotropin levels (hCG) is advocated for those suspected of ectopic pregnancy. This test requires two blood draws separated by a period of 48 hours. 6 Vaginal ultrasound can detect or exclude the presence of an intrauterine sac as early as five weeks after the last m e n s t r u a l period, but the test is highly operator dependent and in m a n y h o s p i t a l s is u n a v a i l a b l e at night and weekends.7, 8 In addition, ultrasound findings often must be correlated with the hCG titer. In 1978, Radwanska et al noted an association between low levels of serum progesterone and abnormal pregnancy. 9 This observation was subsequently confirmed by others in retrospective studies.m, 1] In contrast to serial hCG titers, serum progesterone is not gestational age dependent. Be50/1099
40
GO
80
lOO
cause only a single level is needed, the requirement for a specified time between samples in order to observe a trend in values is eliminated. Following promising results with a progesterone screening program, 12 we sought to compare the value of the clinical history and physical examination to that derived from progesterone screening in a large consecutive series of patients presenting to an emergency d e p a r t m e n t with a high patient incidence of ectopic pregnancy. METHODS This study was conducted in the ED of the Regional Medical Center at M e m p h i s (MED) a 450-bed, acute care h o s p i t a l o w n e d by S h e l b y County Government, operated by a not-for-profit health care corporation and staffed by residents and faculty of the U n i v e r s i t y of T e n n e s s e e , Memphis. As a publicly subsidized facility, the MED provides care to all Shelby County citizens in need, regardless of their ability to pay. The MED ED is the county's busiest and treats more than 67,000 patients annually. Stable "walk-in" patients are triaged to the ED's Acute Ambulatory Care Clinic, where they are evaluated by PGY-1 residents from the DeAnnals of Emergency Medicine
120
partments of Medicine, Surgery, or Obstetrics and Gynecology. Attending supervision is provided at all times. In-house gynecology consultation is also available on request. All reproductive-aged female patients whose last menstrual period was at least 25 days prior to their visit, w h o had a b n o r m a l uterine bleeding or pelvic pain, or when a diagnostic test would be contraindicated in pregnancy had urine collected for hCG testing (Tandem Icon II hCG I m m u n o e n z y m e t r i c Assay, Hybritech Inc, San Diego, California). T h o s e p a t i e n t s p r e s e n t i n g to the MED ED between January 1 and December 31, 1988, and noted to have a positive urine hCG were eligible for inclusion in the study. Pelvic examinations were performed when indicated for gynecologic symptoms (eg, abdominal/pelvic pain, vaginal bleeding or spotting, suspected vaginitis or suspected sexually transmitted disease. )
Serum was also obtained for quantitative analysis of h C G and progesterone levels. If the patient's history and/or physical e x a m i n a t i o n was suggestive of ectopic pregnancy, g y n e c o l o g y c o n s u l t a t i o n was requested and the patient was admitted or discharged at the discretion of this consulting service. 19:10 October 1990
TABLE 1. History and physical examination findings in patients diagnosed with an ectopic pregnancy at their initial ED visit compared with those diagnosed on follow-up
Initial Diagnosis N = 89 (%)
Follow-up Diagnosis N = 72 (%)
P
Vaginal bleeding Amenorrhea
75 (84.3) 48 (53.9)
41 (56.9) 52 (72.2)
NS NS
Pain Cervical motion tenderness Adnexal tenderness Adnexal mass
81 80 82 34
10 6 7 0
History and Physical Examination
(91.0) (89.9) (92.1) (38.2)
(13.9) (8.3) (9.7) (0.0)
< < <
