Incomplete abortion is one of the main causes of maternal mortality and morbidity in the developing world. Infection and hemorrhage are the main complications in incomplete abortion.1 One of the most important aspects in treating incomplete abortion is the sufficient management of pain during the process of evacuating the uterus, regardless of whether patients undergo dilation and curettage with sharp curette or manual vacuum aspiration.2 There is a broad spectrum of pain management, ranging from psychological support in the absence of analgesic medication, to the exclusive use of analgesics, to the use of local anesthesia or general anesthesia.3–5
Evidence suggests that general anesthesia could increase risks for women.6,7 A variety of factors are related to the variations in controlling pain, including institutional norms, the availability of medications, geographical location, provider preference and training for pain management, and the surgical technique used to evacuate the uterus.8,9 In addition, guidelines for appropriate pain management during treatment for incomplete abortion are few and usually not very clear.10,11
The use of sharp curette has been described as being associated with pain in patients undergoing dilation and curettage12; however, in managing incomplete abortion, curettage has been associated with a greater level of pain than manual vacuum aspiration.1 However, there is little information about how useful local anesthesia is when manual vacuum aspiration is used for treating incomplete abortion.
Available studies have evaluated the efficacy of local anesthesia in patients with abortions with a closed cervix, and as such, the evidence cannot be directly extrapolated to the treatment of incomplete abortion, in which the cervix is usually open.13–16 When the cervix is closed, it is necessary to dilate the cervix through mechanical procedures; this is the most painful part of the procedure.17 In an incomplete abortion, the cervix is usually open, requiring no dilation for evacuation. Thus, the role of the paracervical block is less clear.18 It has been reported that the control of pain through the use of pharmacological agents is not necessary when manual vacuum aspiration is being used and that only 10% of patients require medication to lessen their pain.19 Thus, in some hospitals in the Dominican Republic, manual vacuum aspiration is practiced without anesthesia. As such, this study aimed to determine the effectiveness of using paracervical block to control pain among women treated with manual vacuum aspiration for an incomplete abortion with an open cervix compared with no anesthetic pain management.
MATERIALS AND METHODS
The study was conducted in accordance with the protocol and informed consent procedures approved by the Western International Review Board, an external ethical committee fulfilling international regulations, as well as by the institutional review board at Maternidad Nuestra Señora de la Altagracia. The study site Maternidad Nuestra Señora de la Altagracia is a referral hospital for perinatal and maternal care, located in Santo Domingo, Dominican Republic.
An open parallel, randomized clinical trial was designed comparing 2 groups: 1 group, which did not receive any anesthesia, was compared with another group, which received 1% lidocaine paracervical block during manual vacuum aspiration treatment for incomplete abortion with an open cervix.
Patients were assigned to 1 of the 2 groups by means of a random number table, generated by a computational algorithm based on a block size of 2 (20%), 4 (40%), and 6 (40%) to generate a list of treatment allocation. Participants assigned to group 1 would receive paracervical block whereas those assigned to group 2 would not receive any type of local anesthesia before the manual vacuum aspiration procedure. The randomization distribution was kept in sealed, sequential opaque envelopes kept at the Reproductive Health Department’s office at Maternidad Nuestra Señora de la Altagracia and only opened when a study participant had consented to the study and was in the operating theater for treatment.
Women attending Maternidad Nuestra Señora de la Altagracia who were diagnosed as having an incomplete abortion and who fulfilled all the selection criteria (Table 1) were invited to participate in the study. Before treatment, each potential participant received detailed information about the study and was asked to participate voluntarily if she so desired. All participants were asked to sign the study’s informed consent form.
Sociodemographic data, such as age, gestational age, educational level, and social support (understood as referring to help in caring for children or economic help from the husband, a relative, or neighbor), were collected. Gestational age was calculated based on the date of the last period in those cases in which there was certainty regarding this. When there was doubt, this date was estimated by administering a bimanual pelvic examination and/or an ultrasound examination. Low socioeconomic status was ascertained by 2 indicators of basic unsatisfied need: having more than 3 children aged 5 years or less and having no drinking water available.20 Participants also were asked about their number of previous gestations, the timing of the onset of abortion, the presence of pain, the prior use of analgesics or abortifacients, and the desire to be pregnant. For the physical examination, the presence of abdominal pain, uterine size, and the state of the cervix (open or closed) were evaluated. Incomplete abortion was confirmed by an ultrasound examination only when the clinical diagnosis was not clear. All data were registered in the patient clinical chart and in a case report form specifically designed for such purposes.
All the patients received counseling and psychological support before, during, and after the procedure. A social worker was instructed to provide the same psychological support to both study groups, informing the woman about the role of each individual inside the operating room, explaining clearly about what was happening during the entire process, and identifying and responding to the women’s emotional needs. Consenting participants were treated in the operating theater, where the randomization group assignment was opened to determine the pain control to be received. Routine perioperative analgesics were not used in either of the 2 groups. Analgesics were used only in the 6 hours before the procedure when it was strictly necessary because of severe preoperative pain.
The technique for applying the paracervical block was as follows: A 10-mL 1% lidocaine syringe was filled by the physician using an aseptic technique. Per clinical international reproductive health clinical standards, a 23-gauge needle with a needle extender was then used to inject 5 mL of lidocaine slowly to a depth of 0.5 cm in the cervix–vaginal joint at 4- or 5- and 7- or 8-o’clock positions so as to reach the sacrouterine ligaments, the place where sensitive innervations reach the uterus (personal communication, J. Figueroa, 2001).21,22 Five minutes after applying the lidocaine, the gynecologist, using the manual vacuum aspiration technique, evacuated the uterus. Both techniques (paracervical block and manual vacuum aspiration) were standardized with the researcher at Maternidad Nuestra Señora de la Altagracia, who was the only gynecologist conducting both procedures. The number of the cannula used was registered as well as the time the procedure lasted and any complications arising during that time.
During the preoperative period, a trained observer asked the woman to evaluate the degree of anxiety on a visual analog scale (the scale ranged from 0, meaning no anxiety, to 10, meaning maximum anxiety), as well as the level of preoperative pain from 0 (no pain) to 10 (extreme pain).23
Intraoperative pain was evaluated from 2 viewpoints: that of the external observer (on a 0–4 scale) and that of the patient (scale of 0–10) in the postoperative period. The trained observer, who entered the theater just before the manual vacuum aspiration procedure began (thus remaining blinded to the patient’s group assignment), evaluated the patient’s signs of pain by using the following scale:
- 0: Patient did not show any sign of pain.
- 1: Patient showed pain through facial expressions.
- 2: Patient verbally and spontaneously expressed pain
- 3: Patient indicated pain with verbal and facial expressions but allowed the procedure to continue
- 4: Patient did not let the procedure continue without analgesia or anesthetic medication for the pain and demonstrated this by verbal or bodily expression.
Ten minutes after the procedure, the patient was asked to describe the pain that she had been feeling during the manual vacuum aspiration by using the same visual analog scale used in evaluating preoperative pain.
Patients received follow-up care according to standard care procedures. Patients were also evaluated for complications before leaving the institution and on the fifth day after the procedure by appointment. In the absence of problems or complications, participants were released from the study after the 5-day follow-up visit.
The following outcomes were ascertained:
- Intraoperative pain as evaluated by the patient and external observer. The pain expressed by the patient at various points of treatment also was measured both as a discrete variable and as a categorical variable: no pain (0 points) or slight pain (1–3 points), moderate pain (4–6 points), and severe pain (7–10 points).
- Changes in the level of pain on the basis of a comparison of pain before the procedure and pain reported during the procedure. A new variable was created for this quantifying the difference between the 2 point scores.
- The need to suspend the procedure or administer anesthetics medicaments or parenteral sedatives; the need for other parenteral analgesics.
- Existence of intraoperative and postoperative complications (infection, hemorrhage uterine perforation, and incomplete evacuation), the need for further surgery, the presence of adverse events, and the presence of serious adverse events.
A sample size of 215 patients was determined with Power statistical program and confirmed by Tamamut 1.1 statistical program (Pérez A, Rodríguez N, Gil J, Ramirez G, Tamaño de la muestra, Universidad Javeriana, INCLEN, Bogota) when considering a difference between groups of 1.5 points in level of pain measured using a visual analog scale (standard deviation of 3 points), with a level of confidence 0.04, power 90%, and a 1:1 ratio of subjects and considering follow-up loses of 20%. Interim analysis was conducted when half of the patients in the sample had been gathered. To correct the risk of increasing type I errors, the method described by Pockock24 was used.
Patients were evaluated in the groups to which they were assigned according to the analysis of the intention to treat. An interim analysis of the first 108 patients did not reveal significant differences between the 2 groups. No adjustment for covariates was performed in the initial analysis. Stata statistical software (StataCorp, College Station, TX) was used for data entry, cleaning, and for analysis of both the interim and final study data. The biostatistician responsible for processing and analyzing data was blinded regarding the 2 groups being studied.
Student t test and χ2 test were used to compare basic characteristics of the groups. The Wilcoxon signed rank test was used for evaluating the change in pain perceived by patients before and after the intervention; the Mann–Whitney U test was used to determine significance in the differences in the level of postoperative pain experienced between the 2 groups.
The χ2 test was used for comparing how the observer documented pain. The χ2 or Fisher exact test was used to describe complications; the type of complication and the type of procedures requiring anesthetic medications or sedatives were stratified for comparison of the 2 groups.
Statistical modeling was used to evaluate the association between the level of pain during the procedure (no pain or slight pain, moderate pain, and severe pain) and the use of anesthesia, controlling for such confounders as socioeconomic level, prior pain, previous anxiety, age, use of analgesics, time from admission to treatment, and social support. Before applying the modeling, possible colinearities between covariables were evaluated. An ordinal logistic regression model was used to examine the characteristics of the event of interest (an ordinal variable).25 For those variables included in the model and those with more than 2 categories of analysis, dummy variables were defined. In these cases, reference group criteria were defined according to clinical or statistical criteria.
Some interactions were included that were considered important from the clinical point of view and that could have modified the risk. However, none of them proved to be significant, which is why they are not reported in the final model. The ordinal logistic regression model allowed relative risk (RR) and the respective confidence intervals for the intervention group variable to be estimated, adjusted for the possible effect of the confounders.
A total of 215 participants was enrolled in the study between April 2, 2002, and October 23, 2002. Of these, 107 patients were assigned to group 1 and 108 to group 2. One patient entered group 1 with a missed abortion (Figure 1).
The sociodemographic characteristics of the patients studied showed an average age of 25.1 years (standard deviation [SD] ± 5.9) and a mean gestational age of 9.1 weeks (SD ± 1.8). The average weight of patients was 56 kg (SD ± 12.4). The median parity was 2 previous births per patient (range, 0–9). The mean time from admission to treatment was 3 hours (range, 15 minutes to 14 hours). The manual vacuum aspiration procedure lasted an average of 9 minutes (SD 2.5 minutes).
Just fewer than half of the women had attended secondary school (47%); of those, 15% had received university education. The majority (53%) were housewives, but 34% worked outside the home. Although 80% of the patients reported having a steady partner, only 42% of them felt that they had social support. Of the subjects’ dwellings, 15% did not have drinking water available. Three percent of the patients had 3 or more children aged 5 years or less; however, 21% had 2 children aged less than 5 years at the time of the incomplete abortion. Sixteen percent reported monthly family income less than or equal to the current minimum legal wage in the Dominican Republic for 2002 (RD$1,600 = US$80); however, this information was provided by only 71% of the patients. Sixty-five percent of the pregnancies had not been planned.
The most frequently reported medical antecedent was previous surgery, which was reported by 20% of patients; this was followed by allergic and pulmonary illnesses in 6% and 3%, respectively. The vast majority (93%) of patients had had a regular menstrual cycle before the current pregnancy. Almost 6 of 10 women (62%) had had 3 or more gestations with the present pregnancy; 25% had 2 or more children at the time of the study. Spontaneous lower abdominal pain was reported by 86% of the patients; pain was documented upon palpating this region in 54% of patients, and 41% reported pain upon moving the cervix. Pelvic ultrasonography was required for clarifying diagnosis in 82 (38%) patients.
Only 1 patient from group 1 had a closed cervix, which required cervical dilation. A number 12 cannula was used with 50% of the patients and an 8- to 10-diameter cannula with 42% of them.
A comparison of basal characteristics between the 2 groups showed that the groups were similar. However, a greater number of patients in group 1 had a university education (Tables 2 and 3). More than one third (35%) of the patients had taken analgesics 48 hours before the manual vacuum aspiration procedure was performed, but only 3 of these patients had taken them within the 6 hours of the procedure. Before being hospitalized, 32% of the patients from group 1 and 39% of the patients from group 2 had taken some analgesic during the 48 hours before the manual vacuum aspiration (χ2 = 1.51, P = .21). The analgesics used were acetaminophen (26 patients in group 1 and 38 in group 2) and intramuscular sodium diclofenac (8 patients in group 1 and 5 in group 2). Three patients from group 1 had taken ampicillin. Few patients (6%) reported having taken medications with the purpose of inducing an abortion (7 patients in each group).
One in 4 patients reported moderate-to-severe pain in the preoperative period and one quarter indicated having an anxiety level equal to or greater than 3 in the preoperative period. There were no clinically or statistically significant differences between the 2 groups regarding the level of anxiety and preoperative pain (Table 4).
A comparison of the level of pain experienced during the procedure between the 2 groups revealed no statistically significant differences between the pain reported by the patient nor in the degree of pain evaluated by the observer, even though a greater percentage of patients in group 2 was said to be experiencing severe pain, both in the external observer’s evaluation and in the patient’s own evaluation (Table 5). It was estimated that there was a 9% reduction in the risk of severe pain in the group receiving paracervical block. This increase means that for every 11 patients treated with paracervical block, 1 patient did not experience severe pain (crude RR 0.83; 95% confidence interval 0.65, 1.07). However, statistically significant differences were found in each group when comparing the level of preoperative and intraoperative pain described by the patient (Wilcoxon signed rank test; P < .001).
Examination of factors associated with the level of pain revealed that the duration of the manual vacuum aspiration procedure was inversely associated with the level of pain. There was no association between age, educational level, time from admittance to treatment, level of anxiety, preoperative pain, or prior use of analgesics (Table 6).
The model that best explains pain in patients undergoing manual vacuum aspiration took into account preoperative pain and anxiety, the use of paracervical block, and time between admission and manual vacuum aspiration starting; however, none of these variables had a statistically significant association (Table 7). Among those patients who received paracervical block, 89% reported pain at the moment of the injection. None presented nausea, vomiting, or any related allergic reaction.
There was neither the need to suspend treatment nor to administer intraoperative analgesics or intravenous anesthetics for any patient. All evacuations were completed, and no complications were presented. There was no loss to follow-up on the fifth day postoperatively because all patients returned for their schedule appointment as instructed.
Pain control is one of the most relevant aspects in managing incomplete abortion. It has been considered that the intensity of pain caused by manual vacuum aspiration is less than that caused by sharp curettage.1 Our findings suggest that the paracervical block technique used in this study was not significantly more effective for controlling pain than psychological support alone; both were not sufficient for controlling pain from endouterine evacuation using manual vacuum aspiration, and the manual vacuum aspiration technique was associated with severe pain in approximately 50% of the patients.
Operator variability was reduced by standardizing both techniques before initiating the study and by having all procedures conducted by a single physician. However, factors that could have theoretically influenced the low effectiveness of this type of paracervical block include the exact place for injecting lidocaine; the depth to which the anesthetic was injected; the time that elapsed before beginning the procedure; and the amount and type of anesthetic used.
Randomized clinical trials have not found differences with respect to the application of the anesthetic at the 3, 5, 7 and 9 o’clock and the 4 and 8 o’clock positions or circumferential application in the cervix.13,15
With regard to the depth of the local anesthetic injection, Wiebe and Rawling13 found lesser intensity of pain when the injection was performed at a depth of 1.5 inches. However, this technique carries the risk of puncturing the vessels leading to parametrium.
Regarding the amount of anesthetic, in theory tissue distention may put enough pressure on nerves to impair the conduction of pain impulses16; therefore, we used a total of just 10 mL of lidocaine (5 mL in each site) to reduce the distention effect. In terms of the type of anesthetic, the most frequently used is lidocaine, but carbonated lidocaine and chloroprocaine also have been used.13,17 However, some authors consider that the benefit in reducing pain does not come from the medication but from distention, rather, blockage of specific autonomic nerves when there is no waiting period.14
However, with respect to the time spent waiting between the injection and the manual vacuum aspiration procedure, Phair et al26 did not find any differences between starting the manual vacuum aspiration procedure immediately and waiting for 3 to 5 minutes, especially in the endouterine aspiration phase. It is important to state that, in this study, the manual vacuum aspiration procedure was initiated 5 minutes after the block was applied. The previously mentioned studies were performed on patients with closed cervixes; their results are not thus applicable to the target population of the present study.
Another issue to take into account would be the design used. It has been described that in nonblinded studies there is an overestimation of the risk.27 The effect of that bias in the current study would make us reject the null hypothesis. This means the difference between groups would be smaller than mentioned above.
When pain levels are compared with those in previously published studies that evaluated pain exclusively associated with endouterine evacuation, it can be seen that the level of pain ranged from 5.1 to 7.9. This value varies with the use of analgesics in premedication, use (or no use) of local anesthetic, the type of drug used, and the moment when the pain was measured (Table 8).
Donati et al18 evaluated pain by categories, finding 40% of those who had received paracervical block in the moderate-to-severe pain category. Fuentes Velásquez (cited in Memories of the Global Conference on advances and challenges in operational research into postabortion attention), mentioned that approximately 40% of the patients reported severe pain during manual vacuum aspiration.28
It is well known that pain is a biopsychosocial experience, where previous experiences and the sociocultural perception of pain mark great differences in the development, severity, and control of pain.29,30 Thus, measuring the degree of pain becomes a difficult task. This makes it difficult to generalize observations among different populations.
The paracervical block technique used in this study showed a mild but not statistically significant protector effect for severe pain (adjusted RR 0.73) but revealed that there was severe pain associated with manual vacuum aspiration in an important number of patients.
Preoperative pain and preoperative anxiety had a slight effect on the level of pain during the procedure. However, the effect was similarly distributed between the 2 groups. The inverse association between the duration of the manual vacuum aspiration procedure and the pain can be explained by the fact that either the surgeon was performing the technique as quickly as possible to reduce the time the pain lasted or perhaps that in doing the manual vacuum aspiration more quickly there was less delicate handling of the tissues. The paracervical block used in this study and manual vacuum aspiration are safe techniques in that they did not produce any complications.
The paracervical block technique used in this study showed a slight protector effect for severe pain (RR adjusted = 0.73), but there was evidence of severe pain associated with manual vacuum aspiration in a significant number of patients. Psychological support in itself was not enough to control the pain.
Given the fact that no preoperative or intraoperative analgesics were applied in this study, and bearing in mind that the paracervical block was not enough to control the pain, it is recommended that randomized comparative studies be designed to determine the effectiveness of other paracervical block techniques and the efficacy of the use of analgesics in patients presenting with incomplete abortion treated with manual vacuum aspiration.
1. Goyaux N, Alihonou E, Diadhiou F, Leke R, Thonneau PF. Complications of induced abortion and miscarriage in three African countries: a hospital-based study among WHO collaborating centers. Acta Obstet Gynecol Scand 2001;80:568–73.
2. Forna F, Gulmezoglu AM. Surgical procedures to evacuate incomplete abortion [review]. Cochrane Database Syst Rev 2001;CD001993.
3. Girvin S. Increasing access, improving quality: lessons learned from post-abortion care programs, at “Issues in Establishing Postabortion Care Services in Low-Resource Settings” workshop sponsored by the JHPIEGO Corporation in Baltimore, Maryland, May 20–21, 1999. Available at: http://www.reproline.jhu.edu/english/2mnh/2pa/pac_proc/girvin1.htm
. Retrieved March 9, 2004.
4. De Jonge ET, Pattinson RC, Makin JD, Venter CP. Is ward evacuation for uncomplicated incomplete abortion under systemic analgesia safe and effective? A randomized clinical trial. S Afr Med J 1994;84:481–3.
5. Stubblefield PG. Surgical techniques of uterine evacuation in first and second trimester abortion. Clin Obstet Gynecol 1986;13:53–70.
6. Peterson H, Grimes DA, Cates W Jr, Rubin GL. Comparative risk of death from induced abortion under 12 weeks gestation performed with local versus general anesthesia. Am J Obstet Gynecol 1981;141:763–8.
7. Stubblefield PG. Control of pain for women undergoing abortion [review]. Suppl Int J Gynecol Obstet 1989;3:131–40.
8. Lichtenberg ES, Paul M, Jones H. First trimester surgical abortion practices: a survey of National Abortion Federation members. Contraception 2001;64:345–52.
9. Clark S, Krishna U, Kallenbach L, Mandlekar A, Raote V, Ellertson C. Women’s preferences for general anesthesia during first trimester surgical abortion in India. Contraception 2002;66:275–9.
11. Solo J. Easing the pain: pain management in the treatment of incomplete abortion. Reprod Health Matters 2000;8:45–51.
12. Chanrachakul B, Likittanasombut P, O-Prasertsawat P Herabutya Y. Lidocaine verus plain saline for pain relief in fractional curettage: a randomized controlled trial. Obstet Gynecol 2001;98:592–5.
13. Wiebe ER, Rawling M. Pain control in abortion. Int J Gynaecol Obstet 1995;50:41–6.
14. Wiebe ER. Comparison of the efficacy of different local anesthetics and techniques of local anesthesia in therapeutic abortions. Am J Obstet Gynecol 1992;167:131–4.
15. Miller L, Jensen MP, Stenchever MA. A double-blind randomized comparison of lidocaine and saline for cervical anesthesia. Obstet Gynecol 1996;87:600–4
16. Glantz JC, Shomento S. Comparison of paracervical block techniques during first trimester pregnancy termination. Int J Obstet Gynaecol 2001;72:171–8.
17. Cetin A, Cetin M. Effect of deep injections of local anesthetics and basal dilatation of cervix in management of pain during legal abortions: a randomized controlled study. Contraception 1997;56:85–7.
18. Donati S, Medda E, Proietti S, Rizzo L, Spinelli A, Subrizi D, et al. Reducing pain of first trimester abortion under local anesthesia. Eur J Obstet Gynecol Reprod Biol 1996;70:145–9.
19. Leme V, Mtimavalye L, Thole G, Mvula M. The impact of the manual vacuum aspiration (MVA) technique on health care services at Queen Elizabeth Central Teaching Hospital, Blantyre, Malawi. S Afr Med J 1997;87:218–22, 224.
23. Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluating pain. Anesthesia 1976;31:1191–8.
24. Pockock SJ. Clinical trials: a practical approach. Winchester (NY): John Wiley & Sons; 1983.
25. Kleinbaum DG, Klein M. Logistic regression: a self-learning text. Heidelberg, Germany: Springer; 2002.
26. Phair N, Jensen J, Nichols M. Paracervical block and elective abortion: the effect on pain of waiting between injection and procedure. Am J Obstet Gynecol 2002;186:1304–7.
27. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12.
28. Measuring the quality of postabortion attention. In: Advances and challenges in postabortion care operations research. Global Conference Memories, Population Council, New York, 19–21 January 1998. Available at: http://www.popcouncil.org/pdfs/advances_and_challenges.pdf
. Retrieved March 9, 2004.
29. Richards HM, Reid ME, Watt GC. Socioeconomic variations in responses to chest pain: qualitative study. BMJ 2002;324:1308.
© 2004 The American College of Obstetricians and Gynecologists
30. Moore R, Brodsgaard I, Mao TK, Miller ML, Dworkin SF. Perceived need for local anesthesia in tooth drilling among Anglo-Americans, Chinese, and Scandinavians. Anesth Prog 1998;45:22–8.