,
Vol. 26, No. 3, March 1975 Printed in U.SA.
FERTILITY AND STERILITY Copyright "' 1975 The American Fertility Society
MENSTRUAL REGULATION IN THE UNITED STATES: A PRELIMINARY REPORT* WILLIAM E. BRENNER, M.D., DAVID A. EDELMAN, PH.D., AND ELTON KESSEL, M.D.
Department of Obstetrics and Gynecology, and International Fertility l!esearch Program, Carolina Population Center, University of North Carohna, Chapel Hill, North Carolina 27524
Vacuum aspiration of a patient's uterine contents within 14 days of a missed menstrual period to interrupt a possible pregnancy is being done with increasing frequency in the United States. While several names have been given to this practice, "menstrual regulation" has probably been the most freguently used. 1 International reports use different definitions and criteria for this procedure on relatively small numbers of patients. 2-5 Therefore, generalizations from these studies may be inappropriate, but, it appears that (1) menstrual regulation within 14 days is safer than that performed at six weeks; (2) a 4- to 6-mm diameter cannula can effectively interrupt pregnancy; (3) menstrual regulation can be performed on an outpatient basis; and (4) menstrual regulation is performed on many patients who probably do not need it. Whether these interpretations will be consistent with results obtained in this country remains uncertain. The proper vacuum source and pressure, cannula type and size, and criteria for patient selection that will be most safe and effective are not well established. This paper analyzes data on menstrual regulation of 1,009 patients from seven Received April13, 1974. *Supported in part by the International Fertility Research Program of the Carolina Population Center (AID/csd 2979). Presented at the 30th Annual Meeting of the American Fertility Society, April 4 to 6, 1974, Hollywood, Florida.
outpatient clinics in the United States which had recently initiated this service when the study began. MATERIALS AND METHODS
Throughout 1973, the International Fertility Research Program collected data from seven clinics that offered menstrual regulation as a part of their outpatient services. Patient characteristics, medical histories, and other pertinent events were collected by the physicians who performed the procedures. All data were recorded on standard forms. Subjects. Generally, the subjects were young (83.2% under 30 years of age), nulliparous (70% ), and unmarried (75.6%) (Table 1). Follow-up data for 83.3% of the subjects were obtained one to eight weeks after the menstrual regulation procedure . All subjects who were undergoing concurrent surgery, intrauterine device insertion, or who had any known, significant, pre-existing medical condition or prior irregular menstrual periods were excluded from the study. Procedures. All procedures were performed on an outpatient basis according to the same protocol. Medical history was obtained, and physical examination and a pregnancy test (usually Pregnosticon Dri-Dot, Organon, Inc, West Orange, NJ) were performed before the procedure. With the patient in the dorsal lithotomy position, the vagina wa~ cleansed with a bacteriocidal agent. Vacuum aspiration of the uterine contents was performed
289
290
March 1975
BRENNER ET AL
TABLE 1. Selected Patient Characteristics No. of patients
Characteristic
Age (years) 20 20-29 30-39 >39 Mean
20 639 140 29
Parity 0 1-2 3-4 >4 Mean
713 211 71 14
%of
No. of patients
% of patients
772 7 263 12 5
71.6 0.7 26.0 1.2 0.5
patients
19.9 63.3 13.9 2.9
None Analgesia only Local/paracervical block General Unknown
24.3 70.7 20.9 7.0 1.4 0.6
Marital status Single Married Unknown
763 235 11
75.6 23.3 1.1
with either a 4-, 5-, or 6-mm diameter flexible cannula of the Karman type and a modified 50-cc syringe or electrical vacuum pump (Table 2). Analgesia alone or local or paracervical block anesthesia was used (26. 7%) when it was necessary for the patient's comfort (Table 3). If the cannula could not be easily inserted, the cervix was mechanically dilated to a maximum of 7 mm. Cervical dilatation was performed on only 2.8% of the 254 patients for whom data were available. The originally sterile cannula was handled with the "no touch technique." When a smaller cannula was used to stint a larger cannula or a smaller cannula was inserted prior to the study size cannula, it was not considered mechanical dilatation of the cervix. Definitions and criteria. All subjects claimed to be within 14 days of the expected onset of menses. Length of amenorrhea was defined as the number of days TABLE 2. Type of Vacuum Source and Sizes ofCannulae Used Vacuum source
50-cc syringe Cannula size
TABLE 3. Anesthesia Administered Type of anesthesia
Electrical vacuum pump
(mm)
No. of patients
%of patients
No. of patients
%of patients
4 5 6
196 54 10
19.4 5.3 1.0
658 47 44
65.2 4.7 4.4
from the first day of the patient's last normal menstrual period to the day of the menstrual regulation. All subjects were amenorrheic from 29 to 45 days. Pregnancy was documented at the time of the procedure in 54.8% of the patients by the microscopic identification of any one product of conception (chorionic villi, fetal parts, or membranes) in the aspirated uterine contents. Chi-square tests of homogeneity were used to analyze the data; only P values are presented. Menstrual regulation was considered a failure if the subject was still pregnant at the follow-up examination one to eight weeks later. Pregnancy at this time was verified by a positive pregnancy test and artificial abortion results. The failure rate was computed as the number of women still pregnant divided by the number of women pregnant at the time of the menstrual regulation. RESULTS
Effectiveness. Overall, 95.8% of the women with documented intrauterine pregnancies had their pregnancies terminated with the menstrual regulation procedure. Excluding the one woman with a tubal pregnancy, the failure rate was 4.2%; this appeared to be independent of the length of amenorrhea (Fig.l). Although the failure rate did appear to decrease as the duration of amenorrhea increased, the decrease was not significant (P>O.l). When the 50-cc syringe was used, the failure rate was not significantly higher (P>O.l) than when the electrical vacuum source was used for either the 4-, 5-, or
291
MENSTRUAL REGULATION
Vol. 26, No.3
10
8
l
w ~
6
Q::
w
Q::
4
::::> _J
4 u..
2
29-31
25-111
32-33
34-35
36-37
38-39
40-42
43-45
All
84
63
552
LENGTH OF AMENORRHEA {DAYS) N=
6
18
43
137
91
110
FIG. 1. The failure rate in percent (100 x number of patients still pregnant after menstrual regullltion/ number of patients initially pregnant) by duration of amenorrhea. (N=no. of patients documented to be pregnant by identification of chorionic villae). The one patient with an ectopic pregnancy is excluded.
6-mm cannula (Table 4). The lower failure rates with the 4- and 5-mm cannulae and the electrical vacuum pump were not significant (P>O.l). Even when cannula size was not considered, there was no significant difference (P>0.1) between the failure rates of the two vacuum sources. In addition to these variables, the failure rate may be influenced by the experience of the physicians performing the procedures. The number of subjects was too small for an evaluation of all these factors. In some clinics, the failure rate was TABLE 4. Menstrual Regulation Failure Rates by Vacuum Source and Cannula Size Vacuum source 50-cc syringe Cannula size (mm)
No. of patients•
%of
constant over the clinical experience, while in others it was highest during the initial part of the study (Fig.2). If the first 25 cases at each clinic are excluded from the analysis, the failure rate decreases to 3.7%. Complications. The incidence of significant complications was low (Table 5). While the rate was higher among pregnant patients (2.5%) than nonpregnant patients (1.1%), the difference was not significant (P>0.1). The difference appeared to be due to the greater incidence ~ ~
40
-
30
.2..CLI
0
a::: 20 CLI
Electrical vacuum pump
failureb
No. of patients•
%of failureb
4 5 6
96 44 5
5.2 9.1 0
341 29 38
3.5 6.9 0
Total
145
6.2
408
3.4
8 Found to be pregnant at the menstrual regulation procedure. hNumber of patients still pregnant after the menstrual regulation divided by the number pregnant at the menstrual regulation.
....
;::, 10
0
I.L..
0
N = 10 II 13 FIG. 2. The failure rate in percent (100 x number of patients still pregnant after menstrual regulation/ number of patients initially pregnant) o.ver the study period by consecutive patient order numbers from one selected clinic (N=number of patients documented to be pregnant by identification of chorionic villae).
292
BRENNER ET AL
March 1975
TABLE 5. Rate of Specific Complications by Pregnancy Status Pregnant•
Not pregnant• ~
Complications
No.
Cervical laceration (without suture) Hypotension Pelvic infection Prolonged bleeding (>5 days) not requiring curettage Prolonged bleeding requiring curettage Total women with one or more complications
"'
%
No.
%
0.2 0.5 0.4
0 3 2
0.0 0.7 0.4
i~
1
3 2
1.1
3
0.5
0
0.0
14
2.5
5
1.1
roo 90
80 70 60 50 40
lo.
30
"'::;
20
~
~
10
0
25-31
32-33 LENGTH
N=
51
103
~~
30
>-!;;
~~
20
" LENGTH OF AMENORRHEA
of prolonged bleeding in the pregnant patients. None of the subjects had an estimated blood loss greater than 100 ml. Although the complication rates may be dependent upon vacuum source, cannulae size, and length of amenorrhea for both pregnant and nonpregnant women, the small number of subjects in this study with complications did not allow evaluation of all these variables. Selection of patients. Only 54.8% of the patients were actually pregnant. The pro-
~ "'~
40
....""""'"'
0.2
1
•Total number of pregnant patients was 553. bTotal number of patients not pregnant was 456.
lo.
50
1:3~
"'~ 6
~
~~
34-35 OF
186
36-37
38-39
AMENORRHEA
247
40·42
43"45
(CAYS)
198
133
91
FIG. 3. The proportion in percent of the 1,009 women documented to be pregnant (by identification of chorionic villae) at each length of amenorrhea (N =number of patients at each length of amenorrhea).
(DAYS)
FIG. 4. The proportion of positive pregnancy tests in percent by the length of amenorrhea. (N = number of patients documented to be pregnant by identification of chorionic vi].lae).
portion of patients who were pregnant increased progressively after 33 days of amenorrhea (Fig. 3). Only after 35 days of amenorrhea were over one half documented to be pregnant. Generally, the pregnancy test was accurate only when it was positive. However, the proportion of positive tests among patients actually pregnant increased after the 37th day of amenorrhea (Fig. 4). Only after 42 days of amenorrhea was the pregnancy test positive in. more than one half of the pregnant patients. Independent of the duration of amenorrhea, 93.5% of the subjects with a positive pregnancy test were documented to be pregnant, and 97.4% of those not pregnant had negative pregnancy tests. However, the proportion of subjects with a negative pregnancy test who were documented to be pregnant varied from only 39% to 52%, depending upon the duration of amenorrhea. If the pregnancy test was positive, there was a 0.94 probability that the patient was pregnant, independent of the length of amenorrhea (Fig. 5). However, if the pregnancy test was negative, the probability of pregnancy could be determined if the proportion of pregnant patients was known. This proportion may vary from clinic to clinic. Using the known accuracy of the Pregnosticon Dri-
j
293
MENSTRUAL REGULATION
Vol. 26, No.3
1.00
POSITIVE PREGNANCY TEST
0.90
0.80
0.70
1-
z z C)
t:
~
0.30
0.20
~
~ 0.10
ex: a..
0.0 35
31
40
45
LENGTH OF AMENORRHEA (DAYS) FIG. 5. The probability of pregnancy at each length of amenorrhea (number of days from the firSt day of the last normal menstrual period to the day of the menstrual regulation) with either a positive or negative pregnancy test. Calculations were obtained using the known accuracy of the Prognosticon Dri-Dot test. 8
Dot pregnancy test6 and the results of this study, the probability of pregnancy at each length of amenorrhea was calculated (Fig. 5). DISCUSSION
Even in these United States clinics with recently established menstrual regulation services, the procedure was associated with low rates of complications. Of the patients documented to be pregnant, 95.8% had their pregnancies successfully terminated with menstrual regulation. The rates and severity of significant complications among both the pregnant (2.5%) and nonpregnant patients (1.1%) appeared to be clinically acceptable. In contrast to first-trimester abortions performed by vacuum aspiration after seven menstrual weeks' gestation, the complications associated with menstrual
regulation among pregnant women appeared to be less frequent and less severe. Rates of all specific complications for menstrual regulation among pregnant women were low: 0.2% with cervical trauma, 0.5% with neurogenic hypotension, 0.4% with pelvic infection, and 0.5% with prolonged bleeding requiring curettage. Data collected in a similar manner from similar clinics showed that vacuum aspiration between 7 and 12 menstrual weeks' gestation was associated with corresponding rates of 1.1% with uterine injury (cervical lacerations and uterine perforations), 2.4% with pelvic infection, and 1.3% with prolonged bleeding requiring curettage. 5 Failure of menstrual regulation to terminate pregnancy may also be considered a complication. This additional 4.2% complication rate would make menstrual
294
March 1975
BRENNER ET AL
regulation morbidity similar to that of vacuum aspiration at 7 to 12 weeks' gestation. However, only a small proportion of these patients are at risk of severe complications from a subsequent abortion by vacuum. aspiration. Lower failure rates appear feasible since the basic technique of vacuum aspiration is effective. When menstrual regulation was performed by an experienced physician using general anesthesia, the failure rate was less than 1% in gravidas.3 Better physician training, technique, and .patient selection may result in lower failure rates. The number ofsubjects in this study was too small to adequately evaluate variables which may affect the failure rate, such as: vacuum source and pressure or cannulae size and type at each length of amenorrhea, and type of anesthesia and analgesia. Whether or not different or additional procedures, such as a "curette check," would decrease the failure rate without increasing the complication rate is not known. Improved equipment, such as different cannulae designs, may lower the failure rate. Studies are necessary to determine if the complication rate can be lowered and the menstrual regulation procedure be made more comfortable to the patient. Hopefully, when the safest and most effective variables are defined, they may be rapidly instituted in clinical practice. To reduce the proportion of nongravid patients undergoing menstrual regulation, improved methods of patient selection are needed. Unfortunately, the physician has available only the patient's menstrual history, physical examination, urinary pregnancy test, and cervical mucus fern test. Symptoms of pregnancy are usually an unreliable indication of pregnancy prior to seven weeks' gestation. Patients with previous irregular menstrual periods, young patients beginning menarche, older patients approaching menopause, pa-
tients with infrequent or protected intercourse, and patients discontinuing oral contraceptives may have a lower probability of being pregnant. Although the patient can usually report her last normal menstrual period accurately, this information alone is of limited use since we have shown that only after 35 days of amenorrhea are more than 50% of the patients pregnant. If the proportion of pregnant patients in a clinic is similar to that in this study, the probability of pregnancy can be estimated by knowing the amenorrhea length and the pregnancy test results (Fig. 5). For example, if a woman with a positive pregnancy test is 35 days amenorrheic and her pregnancy test is positive, there is a 0.94 probability that she is pregnant. However, if her pregnancy test is negative, there is a 0.45 probability that she is pregnant. It is unknown if consideration of the results of the cervical mucus fern test can further decrease the number of unnecessary procedures. Hopefully, by knowing the probability of pregnancy, the obstetrician can better select patients and decrease the proportion of nongravid patients undergoing the unnecessary hazards and discomforts of menstrual regulation.
..
...
SUMMARY
The efficacy and safety of vacuum aspiration on an outpatient basis without anesthesia or cervical dilatation within 14 days of a missed menstrual period was evaluated in 1,009 women. Among those patients in whom pregnancy could be documented (54.8% }, 4.2% were still pregnant after the procedure. Potentially serious complications were pelvic infection (0.4% ), bleeding requiring a repeat curettage (0.3%), and prolonged bleeding (0.7%). For comfort, 26.7% received local anesthesia or analgesia. To allow selection of a higher proportion of pregnant patients, the probability of pregnancy was derived for patients with either a positive or
•
.
MENSTRUAL REGULATION
Vol. 26, No.3
negative pregnancy test at each length of amenorrhea. To determine the best vacuum pressure and source, and the cannula size and type at each duration of pregnancy, a more extensive (large, controlled, comparative) study is required. REFERENCES 1. Menstrual Regulation-What is It? Population Report, Series F, No.2, April1973 2. Beric B, Kupresanin M, Hulka JF: The Karman catheter: a preliminary evaluation as an instrument for termination of pregnancies up to twelve weeks of gestation. Am J Obstet Gynecol 114(2):273, 1972
295
3. Brenner WE, Edelman DA, Davis GLR, et al: Suction curettage for "menstrual regulation." Proceedings of the 11th Annual Meeting of the American Association of Planned Parenthood Physicians, Houston, Texas, April 11-13, 1973 4. Stirn EM: Minisuction: an office abortion procedure. Proceedings of the 11th Annual Meeting of the American Association of Planned Parenthood Physicians, Houston, Texas, April 11-13, 1973 5. Edelman DA, Brenner WE, Berger G: The effectiveness and complications of abortion by dilatation and vacuum aspiration versus dilatation and rigid metal curettage. Am J Obstet Gynecol119:473, 1974 6. Edelman DA, Brenner WE, Davis GLR, et al: An evaluation of the pregnosticon dri-dot® test in early pregnancy. Am J Obstet Gynecol 119: 521, 1974
Vol. 26, No. 3, March 1975 Printed in U.SA.
FERTILITY AND STERILITY Copyright "' 1975 The American Fertility Society
MENSTRUAL REGULATION IN THE UNITED STATES: A PRELIMINARY REPORT* WILLIAM E. BRENNER, M.D., DAVID A. EDELMAN, PH.D., AND ELTON KESSEL, M.D.
Department of Obstetrics and Gynecology, and International Fertility l!esearch Program, Carolina Population Center, University of North Carohna, Chapel Hill, North Carolina 27524
Vacuum aspiration of a patient's uterine contents within 14 days of a missed menstrual period to interrupt a possible pregnancy is being done with increasing frequency in the United States. While several names have been given to this practice, "menstrual regulation" has probably been the most freguently used. 1 International reports use different definitions and criteria for this procedure on relatively small numbers of patients. 2-5 Therefore, generalizations from these studies may be inappropriate, but, it appears that (1) menstrual regulation within 14 days is safer than that performed at six weeks; (2) a 4- to 6-mm diameter cannula can effectively interrupt pregnancy; (3) menstrual regulation can be performed on an outpatient basis; and (4) menstrual regulation is performed on many patients who probably do not need it. Whether these interpretations will be consistent with results obtained in this country remains uncertain. The proper vacuum source and pressure, cannula type and size, and criteria for patient selection that will be most safe and effective are not well established. This paper analyzes data on menstrual regulation of 1,009 patients from seven Received April13, 1974. *Supported in part by the International Fertility Research Program of the Carolina Population Center (AID/csd 2979). Presented at the 30th Annual Meeting of the American Fertility Society, April 4 to 6, 1974, Hollywood, Florida.
outpatient clinics in the United States which had recently initiated this service when the study began. MATERIALS AND METHODS
Throughout 1973, the International Fertility Research Program collected data from seven clinics that offered menstrual regulation as a part of their outpatient services. Patient characteristics, medical histories, and other pertinent events were collected by the physicians who performed the procedures. All data were recorded on standard forms. Subjects. Generally, the subjects were young (83.2% under 30 years of age), nulliparous (70% ), and unmarried (75.6%) (Table 1). Follow-up data for 83.3% of the subjects were obtained one to eight weeks after the menstrual regulation procedure . All subjects who were undergoing concurrent surgery, intrauterine device insertion, or who had any known, significant, pre-existing medical condition or prior irregular menstrual periods were excluded from the study. Procedures. All procedures were performed on an outpatient basis according to the same protocol. Medical history was obtained, and physical examination and a pregnancy test (usually Pregnosticon Dri-Dot, Organon, Inc, West Orange, NJ) were performed before the procedure. With the patient in the dorsal lithotomy position, the vagina wa~ cleansed with a bacteriocidal agent. Vacuum aspiration of the uterine contents was performed
289
290
March 1975
BRENNER ET AL
TABLE 1. Selected Patient Characteristics No. of patients
Characteristic
Age (years) 20 20-29 30-39 >39 Mean
20 639 140 29
Parity 0 1-2 3-4 >4 Mean
713 211 71 14
%of
No. of patients
% of patients
772 7 263 12 5
71.6 0.7 26.0 1.2 0.5
patients
19.9 63.3 13.9 2.9
None Analgesia only Local/paracervical block General Unknown
24.3 70.7 20.9 7.0 1.4 0.6
Marital status Single Married Unknown
763 235 11
75.6 23.3 1.1
with either a 4-, 5-, or 6-mm diameter flexible cannula of the Karman type and a modified 50-cc syringe or electrical vacuum pump (Table 2). Analgesia alone or local or paracervical block anesthesia was used (26. 7%) when it was necessary for the patient's comfort (Table 3). If the cannula could not be easily inserted, the cervix was mechanically dilated to a maximum of 7 mm. Cervical dilatation was performed on only 2.8% of the 254 patients for whom data were available. The originally sterile cannula was handled with the "no touch technique." When a smaller cannula was used to stint a larger cannula or a smaller cannula was inserted prior to the study size cannula, it was not considered mechanical dilatation of the cervix. Definitions and criteria. All subjects claimed to be within 14 days of the expected onset of menses. Length of amenorrhea was defined as the number of days TABLE 2. Type of Vacuum Source and Sizes ofCannulae Used Vacuum source
50-cc syringe Cannula size
TABLE 3. Anesthesia Administered Type of anesthesia
Electrical vacuum pump
(mm)
No. of patients
%of patients
No. of patients
%of patients
4 5 6
196 54 10
19.4 5.3 1.0
658 47 44
65.2 4.7 4.4
from the first day of the patient's last normal menstrual period to the day of the menstrual regulation. All subjects were amenorrheic from 29 to 45 days. Pregnancy was documented at the time of the procedure in 54.8% of the patients by the microscopic identification of any one product of conception (chorionic villi, fetal parts, or membranes) in the aspirated uterine contents. Chi-square tests of homogeneity were used to analyze the data; only P values are presented. Menstrual regulation was considered a failure if the subject was still pregnant at the follow-up examination one to eight weeks later. Pregnancy at this time was verified by a positive pregnancy test and artificial abortion results. The failure rate was computed as the number of women still pregnant divided by the number of women pregnant at the time of the menstrual regulation. RESULTS
Effectiveness. Overall, 95.8% of the women with documented intrauterine pregnancies had their pregnancies terminated with the menstrual regulation procedure. Excluding the one woman with a tubal pregnancy, the failure rate was 4.2%; this appeared to be independent of the length of amenorrhea (Fig.l). Although the failure rate did appear to decrease as the duration of amenorrhea increased, the decrease was not significant (P>O.l). When the 50-cc syringe was used, the failure rate was not significantly higher (P>O.l) than when the electrical vacuum source was used for either the 4-, 5-, or
291
MENSTRUAL REGULATION
Vol. 26, No.3
10
8
l
w ~
6
Q::
w
Q::
4
::::> _J
4 u..
2
29-31
25-111
32-33
34-35
36-37
38-39
40-42
43-45
All
84
63
552
LENGTH OF AMENORRHEA {DAYS) N=
6
18
43
137
91
110
FIG. 1. The failure rate in percent (100 x number of patients still pregnant after menstrual regullltion/ number of patients initially pregnant) by duration of amenorrhea. (N=no. of patients documented to be pregnant by identification of chorionic villae). The one patient with an ectopic pregnancy is excluded.
6-mm cannula (Table 4). The lower failure rates with the 4- and 5-mm cannulae and the electrical vacuum pump were not significant (P>O.l). Even when cannula size was not considered, there was no significant difference (P>0.1) between the failure rates of the two vacuum sources. In addition to these variables, the failure rate may be influenced by the experience of the physicians performing the procedures. The number of subjects was too small for an evaluation of all these factors. In some clinics, the failure rate was TABLE 4. Menstrual Regulation Failure Rates by Vacuum Source and Cannula Size Vacuum source 50-cc syringe Cannula size (mm)
No. of patients•
%of
constant over the clinical experience, while in others it was highest during the initial part of the study (Fig.2). If the first 25 cases at each clinic are excluded from the analysis, the failure rate decreases to 3.7%. Complications. The incidence of significant complications was low (Table 5). While the rate was higher among pregnant patients (2.5%) than nonpregnant patients (1.1%), the difference was not significant (P>0.1). The difference appeared to be due to the greater incidence ~ ~
40
-
30
.2..CLI
0
a::: 20 CLI
Electrical vacuum pump
failureb
No. of patients•
%of failureb
4 5 6
96 44 5
5.2 9.1 0
341 29 38
3.5 6.9 0
Total
145
6.2
408
3.4
8 Found to be pregnant at the menstrual regulation procedure. hNumber of patients still pregnant after the menstrual regulation divided by the number pregnant at the menstrual regulation.
....
;::, 10
0
I.L..
0
N = 10 II 13 FIG. 2. The failure rate in percent (100 x number of patients still pregnant after menstrual regulation/ number of patients initially pregnant) o.ver the study period by consecutive patient order numbers from one selected clinic (N=number of patients documented to be pregnant by identification of chorionic villae).
292
BRENNER ET AL
March 1975
TABLE 5. Rate of Specific Complications by Pregnancy Status Pregnant•
Not pregnant• ~
Complications
No.
Cervical laceration (without suture) Hypotension Pelvic infection Prolonged bleeding (>5 days) not requiring curettage Prolonged bleeding requiring curettage Total women with one or more complications
"'
%
No.
%
0.2 0.5 0.4
0 3 2
0.0 0.7 0.4
i~
1
3 2
1.1
3
0.5
0
0.0
14
2.5
5
1.1
roo 90
80 70 60 50 40
lo.
30
"'::;
20
~
~
10
0
25-31
32-33 LENGTH
N=
51
103
~~
30
>-!;;
~~
20
" LENGTH OF AMENORRHEA
of prolonged bleeding in the pregnant patients. None of the subjects had an estimated blood loss greater than 100 ml. Although the complication rates may be dependent upon vacuum source, cannulae size, and length of amenorrhea for both pregnant and nonpregnant women, the small number of subjects in this study with complications did not allow evaluation of all these variables. Selection of patients. Only 54.8% of the patients were actually pregnant. The pro-
~ "'~
40
....""""'"'
0.2
1
•Total number of pregnant patients was 553. bTotal number of patients not pregnant was 456.
lo.
50
1:3~
"'~ 6
~
~~
34-35 OF
186
36-37
38-39
AMENORRHEA
247
40·42
43"45
(CAYS)
198
133
91
FIG. 3. The proportion in percent of the 1,009 women documented to be pregnant (by identification of chorionic villae) at each length of amenorrhea (N =number of patients at each length of amenorrhea).
(DAYS)
FIG. 4. The proportion of positive pregnancy tests in percent by the length of amenorrhea. (N = number of patients documented to be pregnant by identification of chorionic vi].lae).
portion of patients who were pregnant increased progressively after 33 days of amenorrhea (Fig. 3). Only after 35 days of amenorrhea were over one half documented to be pregnant. Generally, the pregnancy test was accurate only when it was positive. However, the proportion of positive tests among patients actually pregnant increased after the 37th day of amenorrhea (Fig. 4). Only after 42 days of amenorrhea was the pregnancy test positive in. more than one half of the pregnant patients. Independent of the duration of amenorrhea, 93.5% of the subjects with a positive pregnancy test were documented to be pregnant, and 97.4% of those not pregnant had negative pregnancy tests. However, the proportion of subjects with a negative pregnancy test who were documented to be pregnant varied from only 39% to 52%, depending upon the duration of amenorrhea. If the pregnancy test was positive, there was a 0.94 probability that the patient was pregnant, independent of the length of amenorrhea (Fig. 5). However, if the pregnancy test was negative, the probability of pregnancy could be determined if the proportion of pregnant patients was known. This proportion may vary from clinic to clinic. Using the known accuracy of the Pregnosticon Dri-
j
293
MENSTRUAL REGULATION
Vol. 26, No.3
1.00
POSITIVE PREGNANCY TEST
0.90
0.80
0.70
1-
z z C)
t:
~
0.30
0.20
~
~ 0.10
ex: a..
0.0 35
31
40
45
LENGTH OF AMENORRHEA (DAYS) FIG. 5. The probability of pregnancy at each length of amenorrhea (number of days from the firSt day of the last normal menstrual period to the day of the menstrual regulation) with either a positive or negative pregnancy test. Calculations were obtained using the known accuracy of the Prognosticon Dri-Dot test. 8
Dot pregnancy test6 and the results of this study, the probability of pregnancy at each length of amenorrhea was calculated (Fig. 5). DISCUSSION
Even in these United States clinics with recently established menstrual regulation services, the procedure was associated with low rates of complications. Of the patients documented to be pregnant, 95.8% had their pregnancies successfully terminated with menstrual regulation. The rates and severity of significant complications among both the pregnant (2.5%) and nonpregnant patients (1.1%) appeared to be clinically acceptable. In contrast to first-trimester abortions performed by vacuum aspiration after seven menstrual weeks' gestation, the complications associated with menstrual
regulation among pregnant women appeared to be less frequent and less severe. Rates of all specific complications for menstrual regulation among pregnant women were low: 0.2% with cervical trauma, 0.5% with neurogenic hypotension, 0.4% with pelvic infection, and 0.5% with prolonged bleeding requiring curettage. Data collected in a similar manner from similar clinics showed that vacuum aspiration between 7 and 12 menstrual weeks' gestation was associated with corresponding rates of 1.1% with uterine injury (cervical lacerations and uterine perforations), 2.4% with pelvic infection, and 1.3% with prolonged bleeding requiring curettage. 5 Failure of menstrual regulation to terminate pregnancy may also be considered a complication. This additional 4.2% complication rate would make menstrual
294
March 1975
BRENNER ET AL
regulation morbidity similar to that of vacuum aspiration at 7 to 12 weeks' gestation. However, only a small proportion of these patients are at risk of severe complications from a subsequent abortion by vacuum. aspiration. Lower failure rates appear feasible since the basic technique of vacuum aspiration is effective. When menstrual regulation was performed by an experienced physician using general anesthesia, the failure rate was less than 1% in gravidas.3 Better physician training, technique, and .patient selection may result in lower failure rates. The number ofsubjects in this study was too small to adequately evaluate variables which may affect the failure rate, such as: vacuum source and pressure or cannulae size and type at each length of amenorrhea, and type of anesthesia and analgesia. Whether or not different or additional procedures, such as a "curette check," would decrease the failure rate without increasing the complication rate is not known. Improved equipment, such as different cannulae designs, may lower the failure rate. Studies are necessary to determine if the complication rate can be lowered and the menstrual regulation procedure be made more comfortable to the patient. Hopefully, when the safest and most effective variables are defined, they may be rapidly instituted in clinical practice. To reduce the proportion of nongravid patients undergoing menstrual regulation, improved methods of patient selection are needed. Unfortunately, the physician has available only the patient's menstrual history, physical examination, urinary pregnancy test, and cervical mucus fern test. Symptoms of pregnancy are usually an unreliable indication of pregnancy prior to seven weeks' gestation. Patients with previous irregular menstrual periods, young patients beginning menarche, older patients approaching menopause, pa-
tients with infrequent or protected intercourse, and patients discontinuing oral contraceptives may have a lower probability of being pregnant. Although the patient can usually report her last normal menstrual period accurately, this information alone is of limited use since we have shown that only after 35 days of amenorrhea are more than 50% of the patients pregnant. If the proportion of pregnant patients in a clinic is similar to that in this study, the probability of pregnancy can be estimated by knowing the amenorrhea length and the pregnancy test results (Fig. 5). For example, if a woman with a positive pregnancy test is 35 days amenorrheic and her pregnancy test is positive, there is a 0.94 probability that she is pregnant. However, if her pregnancy test is negative, there is a 0.45 probability that she is pregnant. It is unknown if consideration of the results of the cervical mucus fern test can further decrease the number of unnecessary procedures. Hopefully, by knowing the probability of pregnancy, the obstetrician can better select patients and decrease the proportion of nongravid patients undergoing the unnecessary hazards and discomforts of menstrual regulation.
..
...
SUMMARY
The efficacy and safety of vacuum aspiration on an outpatient basis without anesthesia or cervical dilatation within 14 days of a missed menstrual period was evaluated in 1,009 women. Among those patients in whom pregnancy could be documented (54.8% }, 4.2% were still pregnant after the procedure. Potentially serious complications were pelvic infection (0.4% ), bleeding requiring a repeat curettage (0.3%), and prolonged bleeding (0.7%). For comfort, 26.7% received local anesthesia or analgesia. To allow selection of a higher proportion of pregnant patients, the probability of pregnancy was derived for patients with either a positive or
•
.
MENSTRUAL REGULATION
Vol. 26, No.3
negative pregnancy test at each length of amenorrhea. To determine the best vacuum pressure and source, and the cannula size and type at each duration of pregnancy, a more extensive (large, controlled, comparative) study is required. REFERENCES 1. Menstrual Regulation-What is It? Population Report, Series F, No.2, April1973 2. Beric B, Kupresanin M, Hulka JF: The Karman catheter: a preliminary evaluation as an instrument for termination of pregnancies up to twelve weeks of gestation. Am J Obstet Gynecol 114(2):273, 1972
295
3. Brenner WE, Edelman DA, Davis GLR, et al: Suction curettage for "menstrual regulation." Proceedings of the 11th Annual Meeting of the American Association of Planned Parenthood Physicians, Houston, Texas, April 11-13, 1973 4. Stirn EM: Minisuction: an office abortion procedure. Proceedings of the 11th Annual Meeting of the American Association of Planned Parenthood Physicians, Houston, Texas, April 11-13, 1973 5. Edelman DA, Brenner WE, Berger G: The effectiveness and complications of abortion by dilatation and vacuum aspiration versus dilatation and rigid metal curettage. Am J Obstet Gynecol119:473, 1974 6. Edelman DA, Brenner WE, Davis GLR, et al: An evaluation of the pregnosticon dri-dot® test in early pregnancy. Am J Obstet Gynecol 119: 521, 1974
