Implementation of a piritramide based patient-controlled analgesia (PCA) as a standard of care is not effective for pain control in late abortion induction: a prospective cohort study from a patient perspective
Tascón Padrón L.1, Emrich NLA.1, Strizek B. 1, Gass A.2, Link C.2 Hilbert T. 2 Klaschik S.2, Meissner W.3, Gembruch U.1, Jiménez Cruz J. 1
1 Department of Obstetrics and Prenatal Medicine, University Hospital Bonn, Bonn, Venusberg-Campus 1, 53127 Bonn, Germany.
2 Department of Anesthesiology and Intensive Care Medicine, University Hospital Bonn, Venusberg-Campus 1, 53127 Bonn, Germany
3 Department for Anesthesiology and Intensive Care Medicine/ Department of Palliative Care, University Hospital of Jena,07740 Jena, Germany.
RUNNING HEAD
PCA is inefficient reducing pain in late abortion
ABSTRACT
Objective To assess whether the implementation of patient-controlled analgesia (PCA) with piritramide using an automatic pump system under routine conditions is effective to reduce pain in late abortion inductions
Design Prospective observational cohort study
Setting Patients requiring medically indicated abortion induction from 14 weeks of pregnancy onwards between July 2019 and July 2020 at the department of Obstetrics and Prenatal Medicine of the Bonn University Hospital in Germany.
Methods Evaluation of pain management after implementation of a PCA system compared with previous nurse-controlled tramadol-based standard under routine conditions. Patients answered a validated pain questionnaire and requirement of rescue analgesics was assessed. Pain intensity and satisfaction were measured on a ten-point numeric rating scale.
Main Outcome Measure Maximal pain intensity
Results Forty patients were included. Patients using Piritramide-PCA complained of higher pain sores than those in the standard group (6.90 (± 2.34) vs. 4.83 (± 2.87), (p < 0.05)). In both groups the level of satisfaction with the analgesia received was comparable (8.00 (± 2.45) vs 7.67 (± 2.62), (p = 0.7)). Patients in the PCA group suffered more nausea (63.2% vs 30% respectively, OR 4.0, 95% CI 1.05-15.20, p<0.05) and expressed more the desire for more analgesic support compared to the control group (OR 5.7 (1-33.25), p = 0.05).
Conclusion Women with abortion induction after 14 weeks of gestation suffer from relevant severe pain, which requires adequate therapy. However, addition of PCA does not seem to bring any advantage in patients undergoing this procedure.

Tweetable abstract

Patient-controlled analgesia is not effective as a form of analgesia in medically induction of abortion.
INTRODUCTION
Medically-indicated termination of pregnancy (TOP) is one of the most commonly performed procedures in obstetrics. It is estimated that of the 99,948 TOPs performed in Germany in 2020, almost 3,900 were due to health conditions and a total of 2,874 were performed after the 12th week of pregnancy The standard approach for TOP in the second and third trimester of gestation in Germany is usually medical rather than surgical. This means induction of labour with prostaglandins often after application of mifepristone.
During second or third trimester abortion induction, notably more complications can be expected. In addition to heavy bleeding, nausea and pain, patients may experience higher levels of psychological stress due to longer exposure to pregnancy. It is well established that psychological mechanisms like catastrophizing, resilience and levels of psychological distress play a relevant role in pain modulation. As TOP is an exceptional emotional situation, it is conceivable that this procedure could lead to a higher pain experience in these patients.
Some studies have assessed pain experience and analgetic management in termination of pregnancy in the first trimester . Although little is known about pain management in the later trimesters, it has been shown that suffering pain during birth has a significant long term influence on the physical and psychological wellbeing of mothers even many years after delivery . Being affected by highly intensive pain during delivery has also been correlated with a higher risk of developing postpartum depression Opioid-based patient-controlled analgesia (PCA) is an established technique for postoperative analgesia, has been demonstrated to be an effective method for postoperative pain management and is recommended by evidence-based guidelines worldwide. Patient-controlled remifentanil has been used effectively in the past There is currently no data available for this type of analgesia in medical TOP after the first trimester.
The aim of the present study was to evaluate the introduction of an opioid-based intravenous PCA technique into routine care for pain control during medically induced 2nd or 3rd trimester TOP with the primary outcome measure being maximal pain intensity. Secondary aims were the evaluation of pain management from a patient perspective and evaluation of risk factors for higher pain intensity.
METHODS
This prospective observational study was performed between July 2019 and July 2020 at the department of Obstetrics and Prenatal Medicine of the Bonn University Hospital in Germany. An approval of the local ethics committee was provided (Registration Number: 208/18).
Inclusion criteria were age over 18 years, pregnancy with a gestational age over 14 weeks and medical indication for termination of pregnancy according to the current German regulations. Women with a medical history of opiate abuse or any contraindications for opiate use were excluded from the study.
The procedure of TOP was performed according to standards of the Bonn University Hospital. This consists of taking 600 mg of mifepristone starting in the afternoon on the day of admission, followed by misoprostol the following day at 10:00 pm administered every 4 hours after being informed about off label use of misoprostol for this indication. The dose of misoprostol depends on gestational age and obstetrical history. Patients with a gestational age between 12+0 and 27+6 weeks and without previous uterine operations received a first dose of 200mg orally followed by 400 mg administered vaginally every 4 hours. If the gestational age was over 28+0 weeks, first dose of misoprostol was 100 mg administered orally, increasing to 200 mg of misoprostol every 4 hours vaginally. Patients with a history of uterine surgery prior to abortion induction (such as previous cesarean section) were given a different drug regimen, after being counseled about the potentially increased risk of complications such as uterine rupture. In these cases, patients between 12+0 to 22+6 weeks of gestation received a first dose of 100 mg of misoprostol orally, followed by 100 mg vaginally every 4 hours and patients between 23+0 to 29+6 weeks of gestation were administered 75 mg of misoprostol orally every 4 hours. For those patients receiving TOP later than 31st week of pregnancy after uterine surgery induction of labour was performed by cervical ripening with a Foley-balloon catheter on the same the day as mifespristone administration. The catheter remained in the uterine cavity for 24 hours. If no contractions were achieved after 24 h, oxytocin was administered in the delivery room on the following day after removing the balloon catheter. In cases of more than two previous uterine operations or cesarean sections, or in cases of uterine longitudinal- or T-shaped uterotomy in their medical history, a pharmacological plan for individual abortion induction was prepared.
If TOP was indicated after 20 weeks of gestation, feticide was performed by percutaneous ultrasound-guided injection of potassium chloride into the umbilical vein before administration of mifepristone.
The analgesic management consisted of nurse-controlled administration of ibuprofen 600 mg, paracetamol (acetaminophen) 1 g, metamizole 1 g, tramadol 100 mg or a combination of meptazinol 100 mg and hyoscine butylbromide 40 mg as short infusion on demand depending on patient’s pain. In analogy to analgesic medication for labour in this center no basic analgesia was provided and drug administration was performed depending on pain intensity: nonsteroidal anti-inflammatory drugs (NSAID) for mild pain, tramadol for moderate pain and meptazinol combined with hyoscine butylbromide for severe pain allowing combination in a three-step strategy. For comparison of opioid consumption dosage of the drugs was converted to equivalents of oral morphine using standard literature. According to this it was assumed, that 100 mg meptazinol are equivalent to 12 mg of oral morphine and 15 mg piritramide are equivalent to 30 mg oral morphine.
On April 1st, 2020, an additional form of analgesia was implemented as standard care in our center for women undergoing this procedure. This consisted of intravenous administration of piritramide via PCA pump. A dose of 2 mg of piritramide bolus was given, triggered by the patient as needed. The bolus could be repeated every ten minutes with a maximum dosage of 30 mg in 4 hours. If higher doses were needed, consultation with a specialist from the pain unit was additionally initiated to reevaluate pain management. Otherwise, the patient was seen by a member of the pain unit twice a day. Pump records were evaluated (times and intervals of the applications) and the regime was adjusted as needed. Application of ibuprofen, paracetamol or metamizole was also allowed.
The study group consisted of patients treated after the introduction of the new pain management protocol. The control group was composed of all patients who underwent abortion induction with standard analgesia before the introduction of the PCA pump.
The evaluation was performed according to the QUIPS project (quality improvement in postoperative pain management [23], www.quips-projekt.de, last visited 28th of November 2021) and data collection was performed in a highly standardized manner as described in a previous study. QUIPS is a national multicenter interdisciplinary project for the evaluation of acute pain management using patient reported outcomes (PROs). This is the biggest database for acute postoperative pain worldwide with more than 500,000 patient records in Germany. The pain questionnaire was handed out to the patient at admission. Patients were instructed to fill out the questionnaire anonymously at the time of discharge. Demographical, clinical and outcome data were sampled on the day of discharge. The questionnaire consisted of fifteen questions divided into three thematic groups: pain intensity, impairment due to pain and pain medication and satisfaction with therapy (Table 1). Patients were asked to use a numeric rating scale (NRS) from 0 (no pain) to 10 (most unbearable pain) for questions regarding pain intensity. The NRS was also used for the evaluation of satisfaction with the prescribed analgesia, here being 0 absolutely unsatisfied and 10 very satisfied. The personnel performing the questionnaires and data collection were specifically trained and provided with written guidelines to ensure standardized data collection. Once the questionnaire was filled out by the participants, the data were entered in the central electronic database of the QUIPS project. QUIPS is included in the german registry for clinical trials (DRKS-ID: DRKS00006153)
Pain levels of seven or more were considered severe for the analysis of risk factors for severe pain.
STATISTICAL ANALYSIS
A two-point reduction on the NRS was defined as clinically relevant for the sample size calculation. Based on earlier publications a mean pain score for maximal pain of 4.6 with standard variation of 2.2 was used and nineteen patients were needed in each arm to see a reduction of two points on the NRS with 80 % power and assuming a 5 % type 1 error.
The Mann-Whitney-U-test was used for the analysis of the ordinal variables (NRS for pain intensity and satisfaction) since no normal distribution was observed. The Student’s t-test was performed for continuous variables (dosage of rescue medication, time to first rescue medication). Descriptive categorical variables were analyzed using X2 or Fischer’s exact test as appropriate. Results are expressed as mean, plus / minus standard deviation (SD) for continuous variables and as percentage or odd ratio with 95& confidence interval for categorical variables. Statistical analysis was performed using SPSS version 19.0 (SPSS inc. Chicago, IL).
RESULTS
Forty patients were included in the study. Both the study group and the control group consisted of twenty women each. No statistical differences could be found between both groups concerning demographic data and baseline characteristics (Table 2).
Women in the study group reported higher maximum NRS pain levels (6.90 (± 2.34) vs. 4.83 (± 2.87), (p < 0.05)). Minimal pain and pain related to movement were similar. Satisfaction levels with pain management were similar in both groups (8.00 (± 2.45) vs. 7.67 (± 2.62), (p = 0.7)) (figure 1). The occurrence of nausea was significantly more frequent in patients with piritramide PCA (63.2 % vs 30.0 % respectively, OR 4.0 (95 % CI 1.05 - 15.20) p<0.05). Women in the study group were also more often awake at night due to pain (75 % vs. 20 %, p = 0.048, OR 4.8 (95 % CI 4.2 - 19.9)). No differences between the two groups were observed for dizziness, fatigue or other items of the questionnaire including wanting more analgesia (table 3).
Regarding the use of analgesic drugs, non-opioids were rarely used in both groups (two women received paracetamol in the control group and one woman in the PCA group) except hyoscine butylbromide, since this was administrated in a fixed combination with meptazinol in the control group. Opioid consumption was similar in both groups. Sixteen patients (80 %) in the control group received opioids for analgesia while nineteen women (95 %) used the piritramide pump in the study group (p = 0.34). The analysis of the cumulated dosage of opioid drugs administrated thorough the whole procedure did not significantly differ between the control and PCA group (15 mg (± 13.5) vs. 20 mg (± 20.6) of equivalent oral morphine respectively, p = 0.37).
The analysis of the effect of relevant pain levels shows that patients reporting pain levels of seven or more were more likely to suffer nausea or vomiting (OR 8.5, 95 % CI 1.96 - 36.79, p = 0.004) and dizziness (OR 6.6 (95 % CI 1.62 - 26.87) p = 0.009) and felt more impaired in their mobility (OR 6.79 (95 % CI 1.60 - 28.86) p = 0.009). These patients stated significantly more often that they wanted more analgesics during the whole procedure (OR 5.73 (95 % CI 1.0-33.25), p = 0.05). Feticide performed prior to induction of labour and fetal weight over 350 g showed a significant correlation with the development of severe pain with their distribution being similar. Eleven (64.7 %) patients who had a fetus weighing more than 350g complained about severe pain levels, while only six women (30.4 %) with a fetus of less than 350 g reported maximum pain levels of seven or more. Feticide was also performed in eleven (64.7 %) vs. six (30.4 %) pregnancies. The OR was the same for both variables: feticide and fetal weight over 350 g; 4.19 (95 % CI 1.1-15.9) p = 0.03). No correlation was found between pain intensity higher than seven and any other study variable such as need of curettage, maternal age, induction time, etc.
DISCUSSION
Main findings and Interpretation
In the present study, patients reported high periinterventional pain scores during late medically indicated TOP, even when using patient-controlled analgesia with an automatic intravenous PCA device. To the best of the authors’ knowledge, this is the first study evaluating the use of opioid based PCA for analgesia during TOP in the second and third trimester under standardized conditions. Unfortunately, not much is known about pain, pain management and consequences of pain in patients undergoing medical TOP after the first trimester. The available evidence shows unexpectedly high levels of pain intensity related to this procedure. Therefore, this study group was unable to find any noteworthy evidence, written guidelines or recommendations regarding pain management for women undergoing late medical abortion. Although the TOP procedure itself is subject of heterogeneous regulations in each country and different medical societies have issued different guidelines, pain management is not part of these guidelines.
By introducing PCA as a new standard of care for pain management, the authors expected to observe a reduction of pain levels and an increased satisfaction with pain therapy. However, patients treated with the new standard complained about significantly higher pain levels and these pain levels were related with wish for more analgesia. Other studies came to a similar conclusion when opioids were added to standard pain management in early TOP. Colwill et al. observed that adding oxycodone to standard therapy does not reduce the pain intensity for medical abortion in early pregnancy . In addition, the use of nitrous oxide failed to improve pain management in surgical TOP before 16 weeks of gestation
Several studies have shown that administering opioids as analgesics not only does not reduce pain satisfactorily after surgical TOP but also causes relevant side effects, such as nausea or vomiting. This implies increased discomfort during this stressful process and further medicalization of women who require abortion induction as they need additional antiemetic.
Opioid consumption was over 80 % in both groups, but equivalent cumulative dosages seem to be rather low with 15 mg and 20 mg respectively. However, we have to admit that there is no clear equivalent conversion factor for intravenous meptazinol, since it is an uncommonly used opioid drug outside of obstetrics and currently accepted standard conversion values are based on old studies describing equivalence between intramuscular meptazinol and morphine.
Considering that administration of non-opioid was almost inexistent in both groups, sufficiency of analgesic treatment in both groups can be questioned and considered a plausible explanation for the high levels of pain described in the present study. Since hyoscine butylbromide was always administrated in a fixed combination with meptazinol in the control group, an additional effect of this drug on the observed difference between groups cannot be clearly excluded in this study. Although the analgesic effect of this drug it is not clear, it has been demonstrated, that hyoscine butylbromide reduces the duration of labor, and this could have an effect on the whole pain experience
Nonpharmacological interventions for pain management have been evaluated in other studies. Dufresne et al. studied whether hypnosis before and during TOP had a positive influence on pain management. They showed that patients who were under hypnosis required less intravenous analgesia although they reported similar pain levels compared to women who underwent the standard pain management . The pain questionnaire used for the present study also included one question about the use of non-pharmacological interventions, but patients used these strategies very rarely. For example, relaxation techniques were the most used nonpharmacological method for pain relief, but these were used only by five patients. Although no patients using relaxation techniques reported high severe pain levels, this difference was not statistically significant. Other methods (massage, respiration techniques, local warmth etc.) were used by only one or two patients.
The pain levels reported in this manuscript were unexpectedly high and should be of concern. As a comparison, similar NRS pain levels (6.98 SD +/- 0.27) have been reported after cesarean sections in a recent publication . This is relevant since cesarean section can be considered one of the most painful surgeries compared with non gynaecological procedures and the most painful surgical procedure in the gynaecological spectrum .
Pain scores for operative TOP in the first trimester are not as high as for later medical abortion induction . This can probably be explained by the controlled conditions under which surgical procedures are performed, the use of anesthesia during the process, but also the availability of clear guidelines and recommendations. Patients might also suffer lower mental stress levels due to the shorter duration of pregnancy and different indications for TOP.
In the case of surgical TOP in the first trimester, being married, lower expectations for bleeding and being unafraid of prior pelvic exploration have been related to lower pain scores. In contrast to these data, the present study shows that only fetal weight above 350 g and performance of feticide have a significant correlation with severe pain. Since feticide is only performed in pregnancies after the 20th week of gestation and fetal weigh in these cases is expected to be higher, it is not possible to evaluate the size effect of each variable separately. Other factors such as need of curettage, maternal age or prior obstetric history, did not play a relevant role in this study. However, marital status and patient expectations were not subject of research in the study.
Since medical abortion induces psychological stress and pain modulation is also affected by psychological factors, it could be assumed that these women have a higher perception of pain. This has been already described for catastrophizing and anxiety regarding other medical procedures. Pur et al. for example showed that patients with higher scores for anxiety in the STAI (State-Trait-Anxiety Inventory) questionnaire tended to report higher pain levels after first trimester termination of pregnancy .
Insufficient pain control during abortion procedures is also related with a worse whole health care experience. The present data are in line with these findings since patients who reported high pain scores were less satisfied with the whole management. This has been described for other gynaecological procedures such as cesarean section or gynaecologic laparoscopy .
Recommendations for pain management for second trimester medical abortion have been described as poor and heterogeneous by other groups. Due to the lack of qualitative evidence for analgesic strategies in these patients, each center tends to use local guidelines which are inconsistent and not well evaluated.
Strengths and Limitations
To our knowledge, this is the first study evaluating different pain management modalities in patients undergoing medically induced termination of pregnancy after the first trimester. The strength of this prospective study is the use of a validated high standardized tool for evaluation of acute pain management using patient reported outcomes. Since the purpose of this study was to evaluate the introduction of a PCA pump as a standard of care in daily standard hospital conditions, randomization or blinding was not performed, so selection bias cannot be excluded. In order to address this issue, the following measures were taken: no member of the obstetric team was involved in the data collection as patients filled out the questionnaire alone and external assessors were responsible for data input. Patients were informed about the use of a PCA pump by the anesthesiology team, but they were not aware of the different interventions being evaluated.
The present study has certain limitations. The first of which is the small sample size calculated for the primary outcome. Only nineteen patients per group were needed for each arm to detect a relevant pain reduction with 80 % power. Consequently, this could mean that the evaluation of secondary outcomes is statistically underpowered. Another limitation of this study is the change of setting within the study period. The prenatal unit moved to another building within the hospital nearly four weeks after introducing PCA pumps. The influence of this issue is uncertain, since the personnel and the rest of the standards were unchanged. The study center is a high-volume reference center for prenatal diagnosis and thus no significant fluctuations in the reference population were expected. This could be shown since no significant difference between the groups could be found for demographical characteristics. The study did not evaluate the stress or anxiety levels of the women undergoing TOP. The effect of these factors has been described in other studies, so it is not possible for the authors to rule out the influence of these aspects on the results of the present study.
Conclusion
TOP in the late trimester is associated with high pain intensity. Using piritramide PCA for this procedure was associated with higher pain intensity, nausea and vomiting than the previous standard pain medication on demand. Higher fetal weight and feticide seem to be associated with higher pain levels. Since very little is known about appropriate pain management in these women, more studies are needed to improve patient care.