Frequency of Symptoms of Ovarian Cancer in Women Presenting to Primary Care Clinics
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ORIGINAL CONTRIBUTION
Frequency of Symptoms of Ovarian Cancer
in Women Presenting to Primary Care Clinics
Barbara A. Goff, MD Context Women with ovarian cancer frequently report symptoms prior to diagno-
Lynn S. Mandel, PhD sis, but distinguishing these symptoms from those that normally occur in women re-
mains problematic.
Cindy H. Melancon, RN†
Objective To compare the frequency, severity, and duration of symptoms between
Howard G. Muntz, MD women with ovarian cancer and women presenting to primary care clinics.
O
VARIAN CANCER HAS OF - Design, Setting, and Patients A prospective case-control study of women who
ten been called the “silent visited 2 primary care clinics (N=1709) and completed an anonymous survey of symp-
killer” because symptoms toms experienced over the past year (July 2001-January 2002). Severity of symptoms
was rated on a 5-point scale, duration was recorded, and frequency was indicated as
are not thought to de-
number of episodes per month. An identical survey was administered preoperatively
velop until advanced stages when to 128 women with a pelvic mass (84 benign and 44 malignant).
chance of cure is poor. In fact, text-
Main Outcome Measures Comparison of self-reported symptoms between ovar-
books in internal medicine, family prac-
ian cancer patients and women seeking care in primary care clinics.
tice, and even gynecology state that
symptoms do not occur until the dis- Results In the clinic population, 72% of women had recurring symptoms with a me-
ease is advanced.1-4 However, several dian number of 2 symptoms. The most common were back pain (45%), fatigue (34%),
bloating (27%), constipation (24%), abdominal pain (22%), and urinary symptoms
retrospective studies have indicated that (16%). Comparing ovarian cancer cases to clinic controls resulted in an odds ratio of
the majority of patients do have symp- 7.4 (95% confidence interval [CI], 3.8-14.2) for increased abdominal size; 3.6 (95%
toms, although not necessarily gyne- CI, 1.8-7.0) for bloating; 2.5 (95% CI, 1.3-4.8) for urinary urgency; and 2.2 (95% CI,
cologic in nature.5-9 These studies have 1.2-3.9) for pelvic pain. Women with malignant masses typically experienced symp-
been criticized because of small num- toms 20 to 30 times per month and had significantly more symptoms of higher se-
bers of patients included and the ret- verity and more recent onset than women with benign masses or controls. The com-
rospective chart analyses used for data bination of bloating, increased abdominal size, and urinary symptoms was found in
collection. 43% of those with cancer but in only 8% of those presenting to primary care clinics.
In a previous study, we surveyed 1725 Conclusions Symptoms that are more severe or frequent than expected and of re-
women with ovarian cancer.10 Surveys cent onset warrant further diagnostic investigation because they are more likely to be
were returned from women in 46 states associated with both benign and malignant ovarian masses.
and 4 Canadian provinces. We found JAMA. 2004;291:2705-2712 www.jama.com
that 95% of women with ovarian can-
cer reported symptoms prior to diagno- vanced disease reported symptoms. tient delays (ignoring symptoms) and
sis, with the most common being ab- There was no significant difference in the physician delays (wrong diagnosis, not
dominal (77%), gastrointestinal tract type of symptoms based on early or late- performing pelvic examination, not or-
(70%), pain (58%), constitutional (50%), stage disease. Other investigators have dering radiographic studies, or not de-
urinary (34%), and pelvic (26%). Inter- also shown that 80% to 90% of women termining serum cancer antigen 125
estingly, gynecologic symptoms were the with early stage disease will report symp- levels).10 To date, screening modali-
least common of the major groups of toms for several months prior to diag- ties for asymptomatic women, such as
symptoms. When we evaluated symp- nosis.11-13 serum cancer antigen 125 and trans-
toms by stage of disease, we found that Identification of early symptoms may
have important clinical implications be- Author Affiliations: Department of Obstetrics and Gy-
in contrast to what is published in most necology, University of Washington School of Medi-
textbooks, 89% of women with stage I/II cause 5-year survival for early stage dis- cine, Seattle (Drs Goff and Mandel); Conversations,
ease is 70% to 90% compared with 20% Amarillo, Tex (Ms Melancon); and Virginia Mason
disease reported symptoms prior to their Medical Center, Seattle, Wash (Dr Muntz).
diagnosis and 97% of those with ad- to 30% for advanced-stage disease.14 †Deceased.
Other important findings from our prior Corresponding Author: Barbara A. Goff, MD, Divi-
sion of Gynecologic Oncology, Box 356460, Univer-
study were that advanced disease was sity of Washington School of Medicine, Seattle, WA
For editorial comment see p 2755.
significantly associated with both pa- 98195 (bgoff@u.washington.edu).
©2004 American Medical Association. All rights reserved. (Reprinted) JAMA, June 9, 2004—Vol 291, No. 22 2705SYMPTOMS OF OVARIAN CANCER vaginal ultrasound, have not been tinguish symptoms that occur com- stage of disease. All women signed in- shown to be effective in reducing the monly from those that are more likely formed consent. morbidity or mortality of ovarian can- to be associated with ovarian cancer. Statistical analysis was performed us- cer.15 If there were a way for patients The purpose of this study was to ing SPSS statistical software (version and physicians to recognize early symp- identify the frequency, severity, and du- 10.1, SPSS Inc, Chicago, Ill). Continu- toms of ovarian cancer, then this may ration of symptoms typically associ- ous variables were compared using in- have a favorable impact on survival, ated with ovarian cancer in a popula- dependent tests for 2 groups and analy- even in the absence of accurate screen- tion of women presenting to primary sis of variance with post hoc tests for ing studies for asymptomatic women. care clinics. Comparison was made with more than 2 groups. Categorical vari- Another important case-control study 128 women with ovarian masses who ables were analyzed with 2 (for mul- was reported by Olson et al11 from Me- were surveyed about symptoms prior tiple groups) or Mann-Whitney U (for morial Sloan-Kettering Cancer Center in to surgery and before a cancer or be- 2 groups) and medians were analyzed New York, NY. Women with ovarian nign diagnosis was established. using the Kruskal-Wallis H test. Cor- cancer (n=168) and controls (n=251) relations were performed with Pear- were interviewed about symptoms over METHODS son correlation. P⬍.05 was consid- the previous 6 months. Those with can- Approval for this study was obtained ered significant. cer were interviewed on average 4 to 5 from the institutional review boards at months after diagnosis. These authors the University of Washington and Vir- RESULTS found significant differences in symp- ginia Mason Medical Center, both in Se- In the primary care clinics, 1709 women toms between ovarian cancer patients attle. Women visiting either of 2 pri- completed the survey over the 6-month and controls, with bloating, lack of ap- mary care clinics (Family Medicine and period. Approximately 12000 visits by petite, abdominal pain, fatigue, urinary Women’s Clinic) at the University of women were made. Because a large per- frequency, and constipation occurring Washington were asked to voluntarily cent (30%-50%) were repeat visits significantly more frequently in cases fill out an anonymous survey about the within the 6-month period, it was not than in controls. When the authors symptoms they had experienced over possible to calculate an accurate re- looked specifically at patients with early the past year. Participants were given sponse rate to the survey. Patients were stage disease, they also found that 89% a list of 20 symptoms that are typically instructed not to fill out more than 1 of these patients complained of symp- associated with ovarian cancer (BOX). survey. The median age of the pa- toms prior to diagnosis. For women with These included pain, eating difficul- tients surveyed was 45 years (range, early stage disease, bloating was the most ties, abdominal symptoms, bladder 15-90 years). A total of 430 patients common symptom, followed by gastro- symptoms, bowel symptoms, menses, (25%) made a clinic visit for a general intestinal tract disturbances. sexual intercourse, and constitutional check up, 224 (13%) were visiting only While both our prior study10 and Ol- symptoms. They were asked to rate the for a mammogram, and 1055 (62%) ap- son et al11 suggest that the majority of severity on a 5-point scale, provide the pointments were for specific prob- women with early and late-stage ovar- frequency of symptoms as number of lems. The majority of surveys (78%) ian cancer have symptoms, both of episodes per month, and indicate how were returned from patients visiting the these studies have weaknesses that need long the symptom had been present. In Women’s Clinic. The racial distribu- to be addressed. In both studies, women addition, they were surveyed about age, tion of respondents was 81% white, 6% were surveyed or interviewed months race, parity, education, past medical his- Asian, 5% black, 2% Native American, to years after their diagnosis, making tory, and reason for the clinic visit. Sur- 2% Hispanic, and 4% unknown (not recall bias a significant issue. Another veys were filled out over a 6-month pe- indicated). The highest level of educa- concern is the issue of selection bias. riod ( July 2001-January 2002). tion reported was 8th grade or less for In particular, the control group in the A second group of women about to 1%; between 9th and 11th grade, 2%; Olson et al study was not necessarily have surgery to remove an ovarian or 12th grade or high school diploma, women seeking medical care—most pelvic mass filled out an identical form 12%; 2 years of college, 16%; 4 years were contacted by random dialing or regarding their symptoms over the pre- of college, 33%; graduate degree, 33%; were convenience controls. One of the vious year. These were women who pre- and unknown (not indicated), 3%. major criticisms of both studies has sented for gynecology services at both Sixty-four percent of women had been come from physicians in primary care University of Washington and Vir- pregnant and 52% had delivered chil- who point out that many women who ginia Mason Medical Center. Surveys dren. Regarding medical history, 12% present for routine care frequently do were completed prior to surgery and be- indicated they had hypertension, 4% complain of symptoms that are typi- fore women were aware of the patho- diabetes, 3% heart disease, 4% breast cally associated with ovarian cancer but logical diagnosis (benign or malignant). cancer, 1% endometrial cancer, 12% who do not have the disease. There Surveys were then correlated with sur- thyroid disease, and 7% irritable bowel needs to be an appropriate way to dis- gical pathological characteristics and syndrome (IBS). 2706 JAMA, June 9, 2004—Vol 291, No. 22 (Reprinted) ©2004 American Medical Association. All rights reserved.
SYMPTOMS OF OVARIAN CANCER
Of the women who presented for pri- (P = .02). Women with hypertension,
mary care, 95% reported at least 1 symp- pulmonary disease, cardiac disease, and Box. Questionnaire on
tom in the past year. The most com- prior history of cancer did not report a 20 Symptoms
mon symptoms were back pain (60%), higher number of symptoms com- Typically Associated
fatigue (52%), indigestion (37%), uri- pared with other women in the clinic With Ovarian Cancer
nary tract problems (35%), constipa- population. Have you had any of the following
tion (33%), and abdominal pain (28%). When we evaluated the impact of age symptoms in the past year? If you had
The median number of reported symp- on symptoms, we found that women a symptom, please indicate the se-
toms was 4. The median severity of all with no symptoms were significantly verity, frequency (number of times
symptoms was between 2 and 3. In 72% more likely to be postmenopausal than per month), and duration of the
of cases, women had symptoms that oc- premenopausal (P = .003). All symp- symptom.
curred at least once per month. The most toms were less common as age in- Pain
common recurring symptoms were back creased except for urinary tract symp- Pelvic
pain (45%), fatigue (34%), indigestion toms. There was a significant decrease Abdominal
(28%), constipation (24%), abdominal in severity of symptoms except for uri- Back
pain (22%), and urinary tract prob- nary tract symptoms, which signifi- Eating
lems (16%). The median number of re- cantly increased in severity with age Indigestion
curring symptoms was 2 and the me- (P⬍.001). Unable to eat normally
dian severity for all recurring symptoms There were 128 women with pelvic Nausea or vomiting
was between 2 and 3. masses who completed a survey of Weight loss
Among those who presented to pri- symptoms. Most (70%) of these women Abdomen
mary care clinics, there was no signifi- lived in the western Washington area, Abdominal bloating
cant difference in the type of symp- which is a population area with ap- Increased abdomen size
tom, frequency, severity, or duration proximately 2 million women. The re- Able to feel abdominal mass
between the women who presented to mainder were referred from rural east- Bladder
the Family Medicine Clinic compared ern Washington, Idaho, or Alaska (a
Urinary urgency
with the Women’s Clinic. Women pre- population of approximately 1 mil- Frequent urination
senting for a general check up re- lion women). Eighty-four had benign
ported significantly fewer symptoms in masses (n=74) or tumors of low ma- Bowels
the past year (3 vs 4; P=.001) and fewer lignant potential (n=10). In 44 cases, Constipation
recurring symptoms (1 vs 2; P = .001) women had malignant epithelial can- Diarrhea
than women presenting for a problem cers: 11 with early stage disease and 33 Menses
visit. TABLE 1 shows the most com- with advanced disease. In the women Menstrual irregularities
mon symptoms for those women with benign disease, the median age was Bleeding after menopause
(n = 1011) presenting for a problem 55 years and 95% of the women re- Intercourse
visit. The median number of symp- ported symptoms in the prior year; 67% Pain during intercourse
toms reported by these women was 4 reported recurring symptoms; 8% re- Bleeding with intercourse
and the median number of recurrent ported having symptoms for 6 to 12
Miscellaneous
symptoms was 2. Women with diabe- months; and 19% reported having
Fatigue
tes, thyroid disease, and irritable bowel symptoms for more than 1 year before Leg swelling
syndrome (IBS) had significantly more seeing a clinician. The median num- Other
symptoms than other women in the ber of symptoms was 4 and the me- No symptoms
clinic population. Women with IBS dian number of recurring symptoms
were significantly more likely to have was 2 (n=84). In the group with ma-
fatigue, gastrointestinal tract com- lignancy, the median age was 55 years
plaints and abdominal pain compared and 94% of the women reported symp- toms before seeking medical atten-
with other clinic patients (P⬍.001). The toms in the prior year with 67% hav- tion, 36% had symptoms for 2 months
median number of symptoms re- ing recurring symptoms. The median or less; 24%, 2 to 3 months; 3%, 5 to 6
ported by those with IBS was 6 and the number of symptoms was 8 and the me- months; 8%, 7 to 12 months; and 14%,
median number of recurrent symp- dian number of recurring symptoms more than 1 year.
toms was 4. Women with diabetes were was 4 (n = 44). This number was sig- Table 1 shows a comparison of re-
significantly more likely to have back nificantly higher than the number of ported symptoms for women with ovar-
pain, urinary tract symptoms, consti- symptoms reported in the clinic, IBS, ian cancer and those who presented for
pation, fatigue, and abdominal pain or benign mass population (P = .01). problem visits. A separate comparison
compared with other clinic patients When asked the duration of symp- of women with malignancy and IBS re-
©2004 American Medical Association. All rights reserved. (Reprinted) JAMA, June 9, 2004—Vol 291, No. 22 2707SYMPTOMS OF OVARIAN CANCER
vealed significantly higher percent- vs 49%), increased abdominal size (64% centages of diarrhea (25% vs 66%) and
ages in those with malignancy for pel- vs 32%), and urinary tract symptoms indigestion (36% vs 52%) were found
vic pain (41% vs 25%), bloating (70% (55% vs 33%). Significantly lower per- among those with ovarian cancer.
TABLE 2 shows the odds ratios (ORs)
for symptoms in patients with ovarian
Table 1. Women Reporting Various Symptoms in the Past Year
cancer compared with the other groups.
No. (%) of Women
Women with ovarian cancer were sig-
Ovarian Cancer Clinic Visit nificantly more likely to have pelvic
Symptom (n = 44) (n = 1011)* P Value
pain, abdominal pain, difficulty eat-
Type of pain
Pelvic 18 (41) 264 (26) .02 ing, bloating, increased abdominal size,
Abdominal 22 (50) 301 (30) .006 and urinary urgency compared with
Back 15 (34) 617 (61) .001 women seeking care in primary care
Indigestion 16 (36) 374 (37) .54 clinics. When compared with women
Nausea 6 (14) 224 (22) .15 with IBS, the only significant differ-
Bloating 30 (70) 385 (38) ⬍.001 ences were an increase in pelvic pain,
Increased abdominal size 28 (64) 197 (19) ⬍.001 bloating, abdominal size, and urinary
Fatigue 27 (61) 548 (54) .21 tract symptoms. A comparison of symp-
Urinary tract 24 (55) 323 (32) .002 toms between those women with be-
Constipation 22 (50) 363 (36) .09 nign and malignant ovarian masses is
Diarrhea 11 (25) 329 (32) .25 shown in TABLE 3. Evaluating the com-
Postmenopausal bleeding 1 (2) 36 (4) .56 bination of bloating, increased abdomi-
Menstrual irregularity 8 (18) 260 (25) .22 nal size, and urinary tract symptoms re-
Combination of symptoms vealed that 43% of women with cancer
3† 19 (43) 81 (8) ⬍.001 complained of all 3 symptoms com-
4‡ 12 (27) 44 (4) ⬍.001 pared with only 10% with benign
*Patients did not have irritable bowel syndrome. Excludes patients who presented for routine checkup or mammo-
gram only. masses, 13% with IBS, and 8% of the
†Bloating, increased abdominal size, and urinary urgency. clinic population (P⬍.001). Com-
‡Bloating, increased abdominal size, urinary urgency, and constipation.
pared with cancer cases, the ORs for the
combination of 3 symptoms were 9.4
Table 2. Women With Ovarian Cancer Compared With Those Without Ovarian Cancer for clinic controls, 5.4 for IBS, and 5.3
OR (95% CI) for benign ovarian mass cases (Table 2).
Cancer vs Benign Cancer vs Clinic Cancer vs Correlation of symptoms was evalu-
Symptom Ovarian Tumor Patients* IBS Patients ated for each group of women. In pa-
Type of pain tients with malignancy, bloating was
Pelvic 1.8 (0.8-4.0) 2.2 (1.2-3.9) 2.6 (1.2-5.6)
more highly correlated with increased
Abdominal 1.8 (0.8-4.0) 2.3 (1.2-4.4) 0.7 (0.3-1.5)
abdominal size and urinary urgency
Back 1.4 (0.6-3.3) 0.4 (0.2-0.7) 0.4 (0.2-0.8)
compared with the other groups.
Difficulty eating 2.5 (0.9-6.8) 2.5 (1.3-5.0) 1.5 (0.7-3.7)
Comparison of severity of symptoms
Nausea 0.9 (0.3-2.5) 0.6 (0.2-1.4) 0.6 (0.2-1.4)
among clinic patients and patients with
Weight loss 0.4 (0.1-1.6) 0.7 (0.2-2.1) 0.8 (0.2-3.2)
a benign ovarian mass, ovarian cancer,
Bloating 3.5 (1.5-8.2) 3.6 (1.8-7.0) 3.0 (1.3-6.7)
or IBS revealed that bloating was signifi-
Increased abdominal size 3.0 (1.3-6.9) 7.4 (3.8-14.2) 4.6 (2.1-10.1)
cantly more severe in those patients with
Abdominal mass 1.4 (0.5-3.4) 5.4 (2.4-12.0) 7.4 (2.3-23.5)
benign and malignant ovarian masses
Fatigue 1.1 (0.5-2.6) 1.4 (0.7-2.7) 1.1 (0.5-2.3)
Urinary tract
and those with IBS compared with other
Urgency 3.5 (1.6-8.2) 2.5 (1.3-4.8) 2.6 (1.2-5.7) clinic patients (P=.001). When we evalu-
Frequency 1.9 (0.8-4.3) 1.5 (0.8-2.8) 2.5 (1.2-5.3) ated symptoms with a severity of 4 or
Constipation 3.5 (1.5-8.1) 1.6 (0.9-3.0) 1.0 (0.5-2.2) greater (TABLE 4), we found that women
Diarrhea 1.6 (0.6-3.9) 0.7 (0.4-1.4) 0.2 (0.1-0.5) with ovarian cancer and IBS are signifi-
Menstrual irregularity 1.2 (0.5-3.3) 0.7 (0.3-1.4) 0.7 (0.3-1.7) cantly more likely to have more severe
Combination of symptoms symptoms compared with women with
3† 5.3 (2.2-12.6) 9.4 (5.0-17.7) 5.4 (2.4-12.2) benign masses and other clinic pa-
4‡ 6.2 (2.0-18.8) 8.6 (4.2-17.4) 3.8 (1.5-9.7) tients.
Abbreviations: CI, confidence interval; IBS, irritable bowel syndrome; OR, odds ratio.
*Excludes those patients who presented for routine checkup or mammogram only. Comparison of median frequency of
†Bloating, increased abdominal size, and urinary urgency. symptoms is shown in TABLE 5. Women
‡Bloating, increased abdominal size, urinary urgency, and constipation.
with malignancies have more fre-
2708 JAMA, June 9, 2004—Vol 291, No. 22 (Reprinted) ©2004 American Medical Association. All rights reserved.SYMPTOMS OF OVARIAN CANCER
quent pelvic pain, abdominal pain,
Table 3. Comparison of Benign and Malignant Ovarian Masses*
bloating, fatigue, and urinary tract
Median No. (%) With
symptoms compared with other clinic Total No. (%) P Symptom Recurring P
patients. Women with ovarian cancer Symptom of Women Value Severity Symptoms Value
typically report that symptoms occur Pelvic pain
Benign 31 (37) 3.0 13 (16)
every day compared with clinic pa- .93 .20
Malignant 18 (41) 4.0 13 (29)
tients who typically only have symp-
Abdominal pain
toms 2 to 3 times per month. Interest- Benign 35 (42) 3.0 18 (21)
.21 .44
ingly, women with benign masses have Malignant 22 (50) 4.0 13 (29)
a high frequency of bloating, fatigue, Bloating
and constipation—each of which oc- Benign 41 (49)
.01
3.0 17 (20)
.02
cur almost daily. Malignant 30 (70) 3.0 20 (45)
Evaluation of duration of symptoms Increased abdominal size
Benign 38 (45) 3.0 13 (15)
is shown in TABLE 6. In general, women .02 .33
Malignant 28 (64) 4.0 9 (21)
with both benign and malignant masses
Fatigue
have symptoms of significantly shorter Benign 47 (56) 3.0 21 (25)
.63 .83
duration. For women with malig- Malignant 27 (61) 3.0 11 (24)
nancy, the median duration is 6 months Urinary tract
or less for all reported symptoms. For Benign 26 (31)
.02
3.0 10 (12)
.34
women with IBS or other clinic pa- Malignant 24 (55) 4.0 8 (19)
tients, the median duration of symp- Constipation
Benign 21 (25) 3.0 13 (15)
toms is typically 12 to 24 months. .01 .17
Malignant 22 (50) 3.0 11 (24)
Secondary analysis was performed in
*Symptom severity scale is rated on a scale of 1 to 5, with 1 being minimal and 5 being severe.
women with early (n=11) vs late-stage
(n=33) ovarian malignancy. TABLE 7
Table 4. Symptom Severity of 4 or Higher*
shows the comparison of symptom re-
No. (%) of Women
porting. The small numbers limit the
power of the analysis, but most symp- Benign
Ovarian Cancer Ovarian Mass IBS Clinic P
toms, including bloating, were seen with Symptom (n = 44) (n = 84) (n = 109) (n = 1600) Value
equal frequency between the 2 groups. Type of pain
There were also no differences in fre- Pelvic 16 (36) 13 (15) 10 (9) 160 (10) ⬍.001
quency, severity, or duration; how- Abdominal 10 (23) 9 (11) 22 (21) 121 (7) ⬍.001
ever, larger numbers of cases are re- Indigestion 4 (9) 4 (5) 16 (15) 111 (7) .03
quired to confirm these findings. Bloating 20 (45) 12 (14) 19 (17) 123 (8) ⬍.001
Increased abdominal size 18 (41) 12 (14) 15 (14) 86 (5) .001
COMMENT Fatigue 10 (23) 12 (14) 30 (27) 269 (17) .02
Over the past decade, research efforts Urinary tract 11 (25) 6 (7) 21 (19) 192 (12) ⬍.001
have focused on screening and diagnos- Constipation 9 (20) 6 (7) 19 (17) 106 (7) ⬍.001
tic protocols to detect ovarian cancer Diarrhea 2 (5) 6 (7) 26 (24) 93 (6) ⬍.001
during the early stages. Unfortunately, Abbreviation: IBS, irritable bowel syndrome.
*Symptom severity scale is rated on a scale of 1 to 5, with 1 being minimal and 5 being severe.
attaining this goal has remained elu-
sive, and to date no screening test or sur- Table 5. Median Number of Episodes of Each Symptom per Month
veillance strategy has been shown to re-
Median (IQR)
duce ovarian cancer mortality.15 In a
study evaluating the efficacy of trans- Ovarian Cancer Benign Ovarian Mass IBS Clinic P
Symptom (n = 44) (n = 84) (n = 109) (n = 1600) Value
vaginal sonographic screening in 14469
Type of pain
asymptomatic women at risk for ovar- Pelvic 24 (3-30) 4 (1-29) 2 (1-5) 2 (1-4) .001
ian cancer, Van Nagell et al 16 con- Abdominal 30 (3-30) 7 (4-30) 3 (1-14) 2 (1-5) ⬍.001
cluded that annual transvaginal ultra- Indigestion 18 (5-30) 9 (3-39) 4 (2-23) 3 (2-9) .03
sound examinations are associated with Bloating 30 (4-30) 20 (1-30) 4 (1-18) 2 (1-5) ⬍.001
a decrease in stage at detection and a de- Fatigue 30 (20-30) 28 (8-30) 25 (4-30) 8 (3-29) ⬍.001
crease in case-specific ovarian cancer Urinary tract 30 (23-30) 5 (3-30) 20 (4-30) 12 (3-30) .02
mortality. However, in this study there Constipation 14 (4-30) 25 (6-30) 2 (1-5) 2 (1-5) ⬍.001
were 57214 scans performed, and if the Diarrhea 4 (2-15) 4 (1-30) 3 (1-6) 2 (1-4) .05
6 patients with borderline and granu- Abbreviations: IBS, irritable bowel syndrome; IQR, interquartile range.
©2004 American Medical Association. All rights reserved. (Reprinted) JAMA, June 9, 2004—Vol 291, No. 22 2709SYMPTOMS OF OVARIAN CANCER
losa cell tumors are excluded, there were study, the authors concluded that nation.19 In addition, with limited health
only 11 invasive epithelial cancers de- screening in this population was not ef- care dollars, screening tests need to be
tected, 5 of which were stage I, 3 stage fective in reducing morbidity or mor- cost-effective.
II, and 3 stage III. This translates into tality from ovarian or other mullerian Given that 80% to 90% of women
5200 ultrasounds for each case of inva- cancers. Six of 8 cancers detected dur- who develop ovarian cancer will not
sive cancer detected. A total of 180 ing surveillance were stage IIIC. While have a worrisome family history, and
women in this study underwent sur- studies of serum proteomics may hold screening in the general population is
gery (or 16 surgeries per case of inva- promise for the future,18 currently the not yet effective,15 it is important for
sive cancer). In another study by Liede US Preventive Services Task Force has women and practitioners to under-
et al, 17 screening with transvaginal graded the routine screening of ovar- stand the symptoms of ovarian cancer
sonography and cancer antigen 125 was ian cancer with a “D” ranking, which is so that diagnoses can be made as
undertaken in Ashkenazi Jewish women defined as fair evidence to recommend promptly as possible. Theoretically,
at high risk for ovarian cancer. In this its exclusion in a periodic health exami- prompt diagnoses could lead to detec-
tion at earlier stages when chance of
cure is significantly greater. Even if ear-
Table 6. Median Duration of Each Symptom in Months
lier diagnosis through symptoms does
Median (IQR) not result in detection of earlier stage
Ovarian Cancer Benign Ovarian Mass IBS Clinic P disease, it may allow the performance
Symptom (n = 44) (n = 84) (n = 109) (n = 1600) Value of an optimal cytoreduction in ad-
Type of pain vanced disease. Optimal cytoreduc-
Pelvic 3 (2-9) 2 (1-5) 18 (7-60) 11 (2-18) .001
tion is associated with cure rates of 30%
Abdominal 4 (1-11) 5 (2-9) 12 (2-111) 11 (3-12) .03
Indigestion 6 (2-12) 3 (1-12) 12 (4-120) 12 (4-18) .10
to 40% compared with 0% to 20% for
Bloating 3 (1-6) 3 (1-6) 18 (12-165) 12 (4-12) ⬍.001
suboptimal cytoreduction, and me-
Fatigue 3 (1-6) 5 (2-8) 12 (6-48) 12 (6-24) ⬍.001
dian survival of more than 50 months
Urinary tract 4 (1-10) 5 (1-10) 12 (6-24) 12 (4-24) .01
compared with 36 months.20
Constipation 3 (1-8) 12 (7-12) 24 (12-126) 12 (5-24) .001
Our study, like previous ones,5-12,21
Diarrhea 5 (1-12) 3 (1-7) 21 (12-180) 12 (2-12) .21 has shown that symptoms are com-
Abbreviations: IBS, irritable bowel syndrome; IQR, interquartile range. monly found in women with ovarian
cancer: 94% of patients did have symp-
toms in the prior year and 67% had re-
Table 7. Comparison of Symptoms in Women With Early Compared With Late-Stage curring symptoms. The symptoms most
Ovarian Cancer commonly reported were bloating, in-
No. (%) of Women creased abdominal size, fatigue, uri-
Early Stage Late Stage nary tract symptoms, and pelvic or ab-
Symptom (n = 11)* (n = 33)† P Value dominal pain. These findings are
Type of pain consistent with studies by Olson et al11
Pelvic 7 (64) 11 (33) .09
and our prior study,10 both of which
Abdominal 4 (36) 18 (55) .49
found that abdominal and gastrointes-
Back 5 (45) 10 (30) .46
tinal tract symptoms were the predomi-
Thigh 3 (27) 7 (21) .62
nant complaints in women with ovar-
Indigestion 1 (9) 15 (45) .03
ian cancer. Interestingly, many of the
Difficulty eating 1 (9) 12 (36) .13
women with benign ovarian masses had
Nausea 2 (18) 4 (12) .63
Weight loss 0 5 (15) .31
similar complaints to those with ma-
Bloating 6 (55) 24 (73) .28
lignant masses. Benign masses have the
Increased abdominal size 6 (55) 22 (67) .42
ability to produce significant gastroin-
Abdominal mass 3 (27) 6 (18) .66
testinal tract and abdominal symp-
Fatigue 7 (64) 20 (61) ⬎.99
toms. In a study by Vine et al,13 the au-
Urinary tract thors compared symptoms in 616
Urgency 7 (64) 17 (52) .72 women with ovarian cancer with 151
Frequency 5 (45) 15 (45) ⬎.99 women with ovarian tumors of low ma-
Constipation 5 (45) 17 (52) .73 lignant potential. In that study, symp-
Diarrhea 3 (27) 8 (24) ⬎.99 toms were reported prior to diagnosis
Menstrual irregularity 3 (27) 5 (15) .39 by 92% of women with invasive can-
*International Federation of Gynecology and Obstetrics stage I/II. cer and 86% of women with border-
†International Federation of Gynecology and Obstetrics stage III/IV.
line tumors. The most common symp-
2710 JAMA, June 9, 2004—Vol 291, No. 22 (Reprinted) ©2004 American Medical Association. All rights reserved.SYMPTOMS OF OVARIAN CANCER toms were pelvic discomfort, bowel include severity and frequency of symp- Another important difference be- irregularities, and urinary tract symp- toms. Compared with women in the gen- tween women with ovarian cancer and toms. Although the authors did not eral clinic population, abdominal pain, those without is the number of symp- have access to staging information, one pelvic pain, bloating, constipation, and toms that women experience. In our can assume that the majority of bor- increased abdominal size are signifi- study, the median number of symp- derline tumors were confined to the cantly more severe in women with ovar- toms among women with ovarian can- ovary. Probably any ovarian mass (be- ian cancer. When compared with women cer was 8 compared with 4 in the clinic nign, borderline, or malignant, even if with IBS only, bloating and urinary tract population, and the median number of confined to the ovary) has a high like- symptoms were more severe in women recurring symptoms was 4 compared lihood of producing symptoms. with ovarian cancer. The number of epi- with 2, respectively (P=.01). We also Women who present for care in pri- sodes per month of each symptom was found significant differences in the me- mary care clinics also commonly have significantly greater for women with ovar- dian number of symptoms between symptoms that can be associated with ian cancer; typically they have a symp- those with benign tumors compared ovarian cancer. Ninety-five percent have tom frequency of 15 to 30 times per with malignant tumors. The median had at least 1 symptom in the prior year month. Women with benign tumors have number of symptoms in women with be- and 72% have recurring symptoms. As a similar frequency pattern. For women nign tumors was 4 compared with 8 in women aged, almost all symptoms be- presenting to primary care clinics, symp- patients with ovarian cancer, and the me- came less frequent and less severe, em- toms typically occur 2 to 3 times per dian number of recurring symptoms was phasizing the importance of not attrib- month, often with menses. Even women 2 compared with 4. Olson et al11 also uting symptoms typical of ovarian cancer with IBS typically have symptoms 2 to 3 found significant differences between to the aging process. Not unexpectedly, times per month with the exception of cases and controls. Women with ovar- women who presented for a general fatigue, diarrhea, and urinary tract symp- ian cancer had a mean (SD) of 3.0 (1.8) check up complained of significantly toms. Olson et al11 also asked women if symptoms vs 0.8 (1.3) for the control fewer symptoms than other clinic pa- they experienced their symptoms con- group (P⬍.001). Eltabbakh et al12 found tients. To some extent this may explain tinuously or intermittently. Olson et al no statistical difference in number of some of the differences in ORs for cases found that bloating, fullness, and abdomi- symptoms between those with stage I/II to controls in our study compared with nal pressure were significantly more invasive ovarian cancer compared with the study by Olson et al. In the Olson et likely to be experienced continuously by borderline tumors. However, in that al study, controls were not necessarily women with cancer compared with con- study, symptom information was ab- women who were visiting a physician; trols (62% vs 36%). stracted retrospectively from chart re- therefore, they were probably not rep- Duration of symptoms is another area views. In our study, the combination of resentative of women presenting to pri- that we find significant differences be- bloating, increased abdominal size, and mary care clinics for problem visits. If the tween women with benign or malig- urinary tract symptoms was seen in 43% control group of women in the Olson et nant ovarian masses compared with of women with ovarian cancer com- al study had fewer complaints it would those women who present to primary pared with only 8% of those presenting increase the ORs, which is what was ob- care clinics. In general, women with to the primary care clinic, emphasizing served. In our study, ORs were 7.4 for ovarian masses will have symptoms with the importance of coexisting symp- increasing abdominal size, 3.6 for bloat- a median duration of 3 to 6 months and toms when trying to distinguish symp- ing, 2.5 for difficulty eating, 2.5 for uri- those with IBS or other patients present- toms typical of malignancy compared nary urgency, and 2.2 for pelvic pain. In ing to primary care clinics will have with those without. While ORs for com- the Olson et al study, ORs were 25.3 for symptoms for a median of 12 months to bined symptoms were somewhat bet- bloating; 8.8, difficulty eating; 6.2, ab- 2 years. The study by Olson et al11 also ter than for single symptoms, ORs were dominal/pelvic pain; and 3.5, urinary found that symptoms of short duration not additive, indicating codependence tract symptoms. All had significant con- were significantly associated with ovar- of symptoms. fidence intervals. Although the ORs are ian cancer compared with controls. In While our current study did find that higher in the Olson et al study, it is re- the study by Vine et al,13 median dura- women who present to primary care assuring to see the reproducibility of their tion of symptoms for women with in- clinics frequently have vague symp- findings in our cases and controls; spe- vasive cancer was 2 to 4 months com- toms that can be associated with ovar- cifically that bloating, increased abdomi- pared with 4 to 6 months for women ian cancer, the important difference is nal size, pelvic or abdominal pain, with borderline tumors. These differ- that these symptoms are less severe and difficulty eating, and urinary tract symp- ences were significantly different. In a less frequent when compared with toms are all significantly more com- study by Eltabbakh et al,12 median du- women with ovarian cancer. Typically, mon in women with ovarian cancer. ration of symptoms for women with in- symptoms occur 2 to 3 times per month Other features that distinguish women vasive tumors was 3.4 months and bor- and are often associated with menses, with ovarian cancer from those without derline tumors was 8.0 months (P=.03). which may explain why these vague ©2004 American Medical Association. All rights reserved. (Reprinted) JAMA, June 9, 2004—Vol 291, No. 22 2711
SYMPTOMS OF OVARIAN CANCER
symptoms become less common and less with ovarian cancer, providing valu- Study concept and design: Goff, Melancon, Muntz.
Acquisition of data: Goff, Melancon, Muntz.
severe as women age. In addition, able information for both women and Analysis and interpretation of data: Goff, Mandel,
women with ovarian cancer typically their clinicians. Symptoms that are Muntz.
Drafting of the manuscript: Goff, Melancon, Muntz.
have symptoms of recent onset and have more severe, more frequent than ex- Critical revision of the manuscript for important in-
multiple symptoms that coexist. pected, and of more recent onset war- tellectual content: Goff, Mandel, Muntz.
This study adds further evidence that rant further diagnostic investigation. Statistical expertise: Mandel.
Obtained funding: Goff, Melancon.
ovarian cancer is not a silent disease. These symptoms are more likely to be Administrative, technical, or material support: Goff,
It is important to emphasize that the associated with ovarian masses, many Mandel, Muntz.
Funding/Support: Supported by Ovarian Cancer
majority of women who have symp- of which may be malignant. Research Fund Inc, New York, NY
toms from our list of 20 complaints will Role of the Sponsor: Ovarian Cancer Research Fund
not have ovarian cancer. Nonetheless, Author Contributions: Dr Goff had full access to all Inc had no role in the design and conduct of the study,
of the data in the study and takes responsibility for in the collection, analysis, and interpretation of the data,
this initial study gives better defini- the integrity of the data and the accuracy of the data and in the preparation, review, or approval of the
tion of symptoms typically associated analysis. manuscript.
REFERENCES
1. Ihde DC, Longo DL. Presentations of the patient with 7. Petignat P, Gaudin G, Vajda D, Joris F, Obrist R. 14. Boring CC, Squires TS, Tong T. Cancer statistics.
cancer: solid tumors in adults. In: Fauci AS, Braunwald Ovarian cancer: the symptoms and pathology: the CA Cancer J Clin. 1993;43:7-26.
E, Isselbacher KJ, Wilson JD, Martin JB, Kasper DL, eds. cases of the Cantonal Cancer Registry (1989- 15. NIH Consensus Development Conference State-
Harrison’s Principles of Internal Medicine. 14th ed. New 1995). Schweiz Med Wochenschr. 1997;127:1993- ment. Gynecol Oncol. 1994;55:S4-S14.
York, NY: McGraw-Hill; 1998:360-362. 1999. 16. Van Nagell JR, DePriest PD, Reedy MB, et al. The
2. Young RC. Gynecologic malignancies. In: Fauci AS, 8. Flam F, Einhorn N, Sjovall K. Symptomatology of efficacy of transvaginal sonographic screening in
Braunwald E, Isselbacher KJ, Wilson JD, Martin JB, ovarian cancer. Eur J Obstet Gynecol Reprod Biol. asymptomatic women at risk for ovarian cancer. Gy-
Kasper DL, eds. Harrison’s Principles of Internal Medi- 1988;27:53-57. necol Oncol. 2000;77:350-356.
cine. 14th ed. New York, NY: McGraw-Hill; 1998: 9. Smith EM, Anderson B. The effects of symptoms 17. Liede A, Karlan BY, Baldwin RL, Platt LD, Kuper-
605-611. and delay in seeking diagnosis on stage of disease at stein G, Narod SA. Cancer incidence in a population
3. Clement KD, Connor PD. Tumors of the female re- diagnosis among women with cancers of the ovary. of Jewish women at risk of ovarian cancer. J Clin On-
productive organs. In: Taylor RB, David AK, Johnson Cancer. 1985;56:2727-2732. col. 2002;20:1570-1577.
TA Jr, Phillips DM, Scherger JE, eds. Family Medi- 10. Goff BA, Mandel L, Muntz HG, Melancon CH. 18. Petricoin EF, Ardekani AM, Hitt BA, et al. Use of
cine: Principles and Practice. 5th ed. New York, NY: Ovarian cancer diagnosis: results of a national proteomic patterns in serum to identify ovarian can-
Spinger-Verlag Inc; 1998:916-924. ovarian cancer survey. Cancer. 2000;89:2068- cer. Lancet. 2002;359:572-577.
4. 1999 compendium of selected publications. ACOG 2075. 19. Paley PJ. Screening for the major malignancies af-
Educational Bulletin No. 250. Washington, DC: Ameri- 11. Olson SH, Mignone L, Nakraseive C, Caputo TA, fecting women: current guidelines. Am J Obstet Gy-
can College of Obstetricians and Gynecologists; 1998: Barakat RR, Harlap S. Symptoms of ovarian cancer. necol. 2001;184:1021-1030.
665-673. Obstet Gynecol. 2001;98:212-217. 20. Ozols RF, Rubin SC, Thomas GM, Robboy SJ. Epi-
5. Wikborn C, Pettersson F, Silfversward C, Moberg 12. Eltabbakh GH, Yadev PR, Morgan A. Clinical pic- thelial ovarian cancer. In: Hoskins WJ, Perez CA, Young
PJ. Symptoms and diagnostic difficulties in ovarian epi- ture of woman with early stage ovarian cancer. Gy- RC, eds. Principles and Practice of Gynecologic On-
thelial cancer. Int J Gynaecol Obstet. 1993;42:261- necol Oncol. 1999;75:476-479. cology. 3rd ed. Baltimore, Md: Lippincott Williams &
264. 13. Vine MF, Ness RB, Calingaert B, Schildkraut JM, Wilkins; 2000:981-1057.
6. Wikborn C, Pettersson F, Moberg PJ. Delay in di- Berchuck A. Types and duration of symptoms prior to 21. Igoe DA. Symptoms attributed to ovarian cancer
agnosis of epithelial ovarian cancer. Int J Gynaecol Ob- diagnosis of invasive or borderline ovarian tumor. Gy- by women with the disease. Nurse Pract. 1997;22:
stet. 1996;52:263-267. necol Oncol. 2001;83:466-471. 122,127-128.
2712 JAMA, June 9, 2004—Vol 291, No. 22 (Reprinted) ©2004 American Medical Association. All rights reserved.You can also read
