Hum. Reprod. Advance Access originally published online on July 27, 2006
Human Reproduction 2006 21(12):3235-3240; doi:10.1093/humrep/del278
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Perioperative and post-operative complications of transvaginal ultrasound-guided oocyte retrieval: prospective study of >1000 oocyte retrievals
1 To whom correspondence should be addressed at: Department of Gynaecology and Obstetrics, University of Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, Lübeck 23538, Germany. E-mail: a.k.ludwig{at}web.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: Although transvaginal ultrasound-guided oocyte retrievals (OR) are performed routinely worldwide, there is very little systematic data about its complications. METHODS: We performed a prospective cohort study following the perioperative and post-operative complications of over 1058 ORs. Additionally, we assessed the pain experienced during the OR. RESULTS: A total of 1166 OR were performed during the study period, of which 1058 (90.7%) ORs were included prospectively. Incomplete data meant that the remaining 9.3% were excluded. No complications were caused by sedation or general anaesthesia. Vaginal bleeding was observed in 2.8% of procedures, without any cases of intra-abdominal bleeding. An injury of pelvic structures (a ureteral lesion) occurred in one case. No case of pelvic infection, but one case of unexplained fever, was observed. A severe ovarian hyperstimulation syndrome (OHSS) occurred in 2.7% of cases. Although most patients tolerated the OR well, 3% of patients experienced severe to very severe pain after the OR and 2% of patients were still suffering from severe pain 2 days after the procedure. The pain level increased with the number of oocytes retrieved. About 0.7% of patients required hospitalization for pain treatment. CONCLUSIONS: Patients can be reassured that overall OR is a safe procedure. However, patients have to be counselled about the minor risks of the OR. The literature on complications is reviewed in the article.
Key words: bleeding/complications/oocyte aspiration/oocyte retrieval
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Transvaginal ultrasound-guided oocyte retrieval (OR) for assisted reproduction was first described in 1985 (Wickland et al., 1985
). Because of its simplicity and effectiveness, it has gained widespread popularity and has now become the gold standard for IVF therapy. Nevertheless, despite the advantages, the aspiration needle may injure the adjacent pelvic organs leading to serious complications. Other common complications are bleeding and pelvic infections. OR are usually performed under general anaesthesia or sedation; both may lead to complications. As infertile, but otherwise usually healthy, patients are undergoing this procedure, they have to be counselled thoroughly about the treatment-related complications and risks. Although ORs are performed routinely worldwide, there is very little systematic data about the complications. Besides limited retrospective analyses and one prospective data collection, there are only case reports available in the world literature. Registry data provide the opportunity of large cohorts that cannot be achieved by prospective studies. However, registries are dependent on the compliance of the centres to report relevant data. Most registries focus on the cycles’ outcomes. Complications of the treatment have been rather neglected so far. As they are not directly associated with the cycle’s outcome, the reporting centres might be less compliant in reporting complications. Especially as complications often occur days to weeks after the OR, the discipline of reporting these complications to the registry might be suboptimal.
The European IVF-Monitoring Programme of the European Society of Human Reproduction and Embryology (ESHRE) analysed 235 324 IVF and ICSI cycles performed in 23 European countries in 2001 (Andersen et al., 2005). Bleeding, without further definition, occurred in 394 cycles (0.17%). Eighty-two per cent (323/394) of these cases were reported in Germany, leading to a bleeding rate of 0.54% in this country. All other countries reported 0–21 cases of bleeding. An infection was reported after 24 cycles resulting in an infection rate of 0.01%. Most countries reported no cases of infection. The large differences between the European countries in bleeding and infectious complications seem to be rather because of different compliances in reporting complications and different designs of the national registries and might also be caused by differences in definitions. However, the low complication rates are questionable.
The German IVF registry included 379 563 ORs from 2000 to 2004. In 28.23% of ORs, no information about complications was available. In those cycles for which information were available, a vaginal bleeding occurred after 0.7% of ORs, an intra-abdominal bleeding in 0.05% of procedures, an injury to the bowel in 0.001% and a peritonitis in 0.005%. An operation was necessary in 0.002% of cycles and a hospitalization in 0.019% of cycles (Deutsches IVF Register, 2005).
Compared with the retrospective and few prospective studies available, these numbers seem to be extremely low. Therefore, we assessed the complications of >1000 ORs prospectively in a standardized manner to obtain reliable data for the consultation of patients.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Patients
All ORs performed at the Division of Reproductive Medicine at the University of Lübeck and at the Endokrinologikum Hamburg from 1 March 2004 to 31 August 2005 were included in the study and followed prospectively. The local ethical review board approved this study.
OR
According to the patients’ preferences, the ORs were performed either under general anaesthesia or under sedation with midazolam (Dormicum® 7.5 Lacktabletten, Hoffman-La Roche AG, Grenzach-Wyhlen, Germany) at a dose of 2.5 mg and pethidine (Dolantin® 50 mg Injektionslösung, Aventis Pharma Deutschland GmbH, Bad Soden, Germany) at a dose of 50 mg.
Before OR, the vagina was soaked with an isotonic saline solution. The patient was covered with a sterile surgical sheet, and the gynaecologist wore sterile surgical gloves. The ultrasound transducer was covered with a sterile plastic sheet (University of Lübeck) or was disinfected in sterile solution for 10 min (Endokrinologikum Hamburg). Preventive antibiotics were not used. The OR was performed under ultrasound guidance with a 16G single-lumen needle. In case of vaginal bleeding after the OR, local compression was applied. If compression of longer duration did not stop the bleeding, then a tamponade was applied for 2 h.
After OR, the patient was kept on the ward for observation for 2 h or as necessary.
Embryo transfer and definition of clinical pregnancy
Cycles were performed under the conditions of the German embryo protection law. No more pronuclear stage oocytes than the number intended for transfer were cultured beyond the pronuclear stage. A maximum of three embryos were transferred; in patients younger than 35 years of age, no more than two embryos were transferred. Embryo transfer was performed 2–3 days after OR.
A clinical pregnancy was defined as a vital pregnancy with a heart beat proved by ultrasound.
Study protocol
All patients undergoing OR at the Division of Reproductive Medicine at the University of Lübeck and at the Endokrinologikum Hamburg signed an informed consent and agreed to be contacted by telephone 2 months after OR for the final evaluation of complications.
No further intervention other than the OR was performed to the patients during the course of this study.
Complications during anaesthesia (need for post-operative observation of >2 h, inadequate oxygen saturation, somnolism and hypotension) or post-operative observation at the intermediate care unit were recorded directly after the OR.
Cases of bleeding were recorded on an evaluation sheet by the performing doctors directly after the OR. It was specified whether a compression of >1 min or a suture was necessary or whether a tamponade had to be administered vaginally to stop the bleeding.
The patients were asked to rank their amount of pain 2 h after oocyte aspiration and directly before the embryo transfer 2–3 days after OR on a scale from 1 to 5, 1 meaning no pain and 5 meaning very severe pain.
Two months after the OR, the patients’ charts were reviewed for the cycle’s outcome and for complications occurring after the retrieval. Additionally, all patients were contacted by telephone 2 months after OR and asked for complications, hospital admissions or any episodes of unexplained fever.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
A total of 1166 ORs were performed during the study period at both IVF centres.
About 1058 (90.7%) of ORs were included prospectively. In 108 cycles (9.3%), the patients had signed an informed consent, but the evaluation sheets recording complications during anaesthesia, bleeding and the post-operative pain score were not available for evaluation. These 108 cycles without the perioperative complications were analysed separately for post-operative complications by chart review and telephone contact (retrospective data). The patients who were followed retrospectively did not differ from those who could be included prospectively. Three hundred and forty of the ORs included in the study were performed at the Division of Reproductive Medicine at the University of Lübeck and 718 at the Endokrinologikum Hamburg. Patients’ characteristics are presented in Table I.
|
In the following, only the prospectively assessed procedures are evaluated. Information on the retrospectively assessed cycles is presented in Tables I and II.
|
Most patients (94.9%) chose general anaesthesia for the OR. The cycles’ characteristics are presented in Table II.
There were no complications from general anaesthesia. In 2.7% of cases, bleeding occurred that required local compression of >1 min. In one case (0.1%), bleeding stopped only after applying a tamponade for 3 h. In none of the cases was a suture necessary (Table III).
|
Two hours after the OR and on the day of embryo transfer 2–3 days after the retrieval, the patients recorded a mean pain score of 1.85 and 1.48, respectively. About 38.5% of patients had no pain 2 h after OR, 41.6% had mild pain, whereas 16.9% of patients recorded medium pain, 2.7% severe pain and 0.4% very severe pain (Figure 1). The mean pain score was dependent on the number of oocytes retrieved. Patients with 6–10 oocytes experienced significantly more pain than patients with 0–5 oocytes (P = 0.003) and significantly less pain than patients with >10 oocytes (P = 0.029) (Figure 2). On the day of embryo transfer, 60% of patients felt no pain from the OR anymore and 25.6% still suffered from mild pain. However, even 2–3 days after OR, 7.6% of patients suffered from medium pain, 1.5% from severe pain and 0.2% from very severe pain, according to their personal self-estimation. Seven patients (0.7%) presented at the emergency room during the days after the OR because of heavy abdominal pain other than ovarian hyperstimulation syndrome (OHSS)-related symptoms were admitted to the hospital.
|
) and before embryo transfer (
). 1, no pain; 2, mild pain; 3, moderate pain; 4, severe pain; 5, very severe pain.
|
In one case, a ureteral lesion occurred during OR. The patient was admitted to the hospital for abdominal pain and vaginal bleeding. After an insertion of a stent into the ureter, the patient recovered quickly and received her embryos back.
None of the 1166 ORs was complicated by clinically apparent pelvic infections. One patient had suffered from unexplained fever 2 weeks after the retrieval that disappeared without therapy.
Severe OHSS occurred in 2.7% of cases.
About 8.1% of patients were admitted to the hospital within 2 months of the OR (Table IV). The most common reasons for hospitalization were complications during early pregnancy (abortion, extrauterine pregnancy and bleeding) in 48 patients (57.1% of those hospitalized) and severe OHSS in 17 patients (20.2% of hospitalized patients). Seven (8.3%) patients were hospitalized because of severe abdominal pain and one patient because of the ureteral lesion. Eleven (13.1%) patients underwent surgery within 2 months of OR because of findings during the last treatment cycles to optimize the chances to conceive (Table IV). Excluding those patients who were hospitalized for complications during early pregnancy, 3.4% (36/1059) of patients undergoing OR were hospitalized within 2 months of the OR for complications of ovarian stimulation, OR or findings during treatment.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
In a prospective study of 1058 ORs, we were able to show that this procedure is safe and well tolerated by the patients. This is the first prospective study following the complications during the OR and the post-operative interval of 2 months in a standardized manner.
Overall, patients tolerated the pain caused by the OR well. The pain increased with the number of oocytes retrieved indicating the greater trauma with a higher number of oocytes. Taking into account that severe stimulation protocols are accompanied by major health risks (Ludwig et al., 1999), the increased pain underlines the necessity to keep ovarian stimulation as soft as possible.
However, a few patients experienced severe pain that sometimes even required hospital admission. Two hours after the procedure, 16.9% of patients rated their pain to be medium and 3.1% to be severe or very severe. Two days after the OR, 7.6% of patients still experienced medium pain and 1.7% severe or very severe pain. In seven cases (0.7%), the patient’s pain required a hospital admission. To date, the issue of pain after OR seems to have been neglected as we were unable to find any data in the literature. Furthermore, there are no data available, which can relate the experience of pain during OR and the subsequent days to the success of a treatment cycle of clinical pregnancy rates. However, there is some evidence that psychological stress exerts a negative influence on the cycle outcome (Klonoff-Cohen et al., 2001).
No complications of sedation or general anaesthesia were observed. In the literature, there are no data on complications during anaesthesia for ORs. One case report describes a patient with a history of heart conduction disease who experienced a severe bradycardia and bradypnoea 85 min after the administration of 400 mg mepivacaine paracervically for a vaginal OR (Ayestaran et al., 2000).
The most common problem associated with OR is minor vaginal haemorrhage that occurred in 2.8% of cases. No case of severe intra-abdominal bleeding was observed in our study. The only other prospective study on this topic observed a vaginal bleeding in 8.6% of ORs (Bennett et al., 1993). However, in most cases, the blood loss was minor; in only 0.8% of ORs, the bleeding exceeded 100 ml and in 1.0% application of local pressure was required. In 0.11% of ORs, a severe intra-abdominal bleeding occurred. The definition of vaginal bleeding as well as the assessment of the amount of blood loss remain unclear in the study by Bennett et al. (1993). Retrospective studies have reported the incidence of severe intra-abdominal bleedings with 0.08–0.2% (Dicker et al., 1993; Tureck et al., 1993; Govaerts et al., 1998) (Table V). Bergh and Lundkvist (1992) interviewed 12 IVF centres about the complications of 10 125 ORs by a questionnaire. They reported that vaginal bleeding (without further specification) occurred in 0.5% of procedures. It is unclear how the reporting centres evaluated their cycles with respect to the complications and how a vaginal bleeding was defined (Bergh and Lundkvist, 1992). In a recent register linkage study on 9175 IVF cycles in Finland, Klemetti et al. (2002) reported bleeding necessitating hospital care in 0.09%. Most instances of bleeding occurred within 2 months of starting therapy. However, the authors do not give information whether these cases of bleeding were a consequence of the OR or whether they were because of hormonal chances or complications of early pregnancy. The start of therapy was defined by the days of the purchase of the medications. Therefore, it is not clear whether the authors really assessed the complications caused by the OR.
|
Severe bleeding can also occur as a consequence of bleeding disorders and medical problems influencing coagulation. Battaglia et al. (2000) and El-Shawarby et al. (2004)
reported on two cases of severe bleeding in patients with a deficit of coagulation factor IX (Battaglia et al., 2001) and in a patient with essential thrombocythaemia (El Shawarby et al., 2004).
All these studies, including the data from the ESHRE and the German IVF registry, as discussed above (Andersen et al., 2005; Deutsches IVF Register, 2005), demonstrate clearly that only prospective studies can report reliable data on those specific questions.
Pelvic structures may be inadvertently traumatized by the aspiration needle. The risk of injury of the bowel appears to be more theoretical than actual. Neither in our study nor in the cohort published by Bennett et al. (1993), an injury of the bowel was observed. Ultrasound guidance allows visualization of the bowel. However, bowel injuries might also occur more frequently without being diagnosed and resolve spontaneously.
Two cases of perforated appendicitis following OR have been published in which puncture holes were found in the appendix (Akman et al., 1995; Roest et al., 1996). Damage of the ureter and ureteral obstruction, as in our study, has also been described in case reports (Jones et al., 1989; Coroleu et al., 1997; Fugita and Kavoussi, 2001; Miller et al., 2002). Given the ureters’ anatomical position immediately anterolateral to the upper fornices of the vagina, it is surprising that the clinical recognizable ureteral injuries do not occur more often than reported. Damage to the bowel seems to be very rare or might cause only minor problems and might therefore remain undetected. We are not aware of any data in the literature on injuries of the bowel during OR besides the cases of perforated appendicitis mentioned above.
Pelvic infections are another rare complication of OR. We did not observe a single case of infection after OR in our cohort. However, the incidence of pelvic infections and tubo-ovarian abscesses reported in the literature is 0.3–0.6%. In a recent review, El-Shawarby et al. (2004) described three different pathways for such infections. Direct inoculation of vaginal micro-organisms may occur by puncture through the non-sterile vagina. The risk of pelvic infections after OR seems also related to the history of pelvic inflammatory disease. Reinfection may occur through puncture of chronically infected ovaries. In a retrospective study, Dicker et al. (1993) described nine cases of tubo-ovarian and pelvic abscesses, all with a previous history of pelvic inflammatory disease. Although least likely, infection may occur through direct puncture of the bowel with an inflammatory or infectious spillage. Bennett et al. (1993) found in their series a minor pelvic infection in 0.3% of cases defined by pyorexia and pelvic tenderness with no evidence of abscess formation on ultrasound. All patients were treated with antibiotic therapy. More severe infections leading to an abscess also occurred in 0.3% of cases (Bennett et al., 1993). This incidence is in accordance with the retrospective analyses (Ashkenazi et al., 1994) (Table VI). However, the preventive treatment with antibiotics does not seem to be helpful and is discussed to be a possible harm for the outcome of IVF cycles.
|
Although only one patient was hospitalized for a severe complication of the OR (injury of the ureter), 36 patients (3.4%) were hospitalized within 2 months after the OR because of complications of ovarian stimulation or OR, pain or findings during treatment that are related or relevant to infertility treatment. This means that 3 in 100 otherwise healthy patients undergoing OR become hospitalized ‘ill’ patients. Although this is no direct complication of the OR, it underlines the consequences of infertility treatment.
In conclusion, patients can be reassured that OR is a safe procedure. However, patients have to be counselled about the possible risks of the OR. The most common complication is clinically relevant vaginal bleeding occurring in 2–3% of procedures. These complications can be treated by local compression; rarely, application of a tamponade or a suture is necessary. Severe intra-abdominal bleeding seems to occur in <1 in 1000 procedures. Although we did not observe a case, infectious complications such as pelvic infections or pelvic abscesses seem to be the second most common complications with 2–6 cases per 1000 ORs according to the literature. Injury of pelvic structures, such as the ureter, the bowel or the appendix, happens in 1 in 1000 procedures. General anaesthesia and sedation are both very safe, with only one case report of a complication after sedation for OR being published in the literature.
Although most patients tolerate the OR well, 3% of patients experience severe to very severe pain and 2% of patients still suffer from severe pain 2 days after the procedure. The pain level increases with the number of oocytes retrieved.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank the teams of the infertility units of the Endokrinologikum Hamburg and of the University of Lübeck for their help with data collection and recording.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Akman MA, Katz E, Damewood MD, Ramzy AI, Garcia JE. (1995) Perforated appendicitis and ectopic pregnancy following in-vitro fertilization. Hum Reprod 10:3325–3326.
Andersen AN, Gianaroli L, Felberbaum R, de Mouzon J, Nygren KG. (2005) Assisted reproductive technology in Europe, 2001. Results generated from European registers by ESHRE. Hum Reprod 20:1158–1176.
Ashkenazi J, Farhi J, Dicker D, Feldberg D, Shalev J, Ben Rafael Z. (1994) Acute pelvic inflammatory disease after oocyte retrieval: adverse effects on the results of implantation. Fertil Steril 61:526–528.[Web of Science][Medline]
Ayestaran C, Matorras R, Gomez S, Acre D, Rodriguez-Escudero FJ. (2000) Severe bradycardia and bradypnea following vaginal oocyte retrieval: a possible toxic effect of paracervical mepivacain. Eur J Obstet Gynecol Reprod Biol 91:71–73.[CrossRef][Web of Science][Medline]
Battaglia C, Regnani G, Giulini S, Madgar L, Genazzani A. (2001) Severe intraabdominal bleeding after transvaginal oocyte retrieval for IVF-ET and coagulation factor XI deficiency: a case report. J Assist Reprod Genet 18:178–187.[CrossRef][Web of Science][Medline]
Bennett SJ, Waterstone JJ, Cheng WC, Parsons J. (1993) Complications of transvaginal ultrasound-directed follicle aspiration: a review of 2760 consecutive procedures. J Assist Reprod Genet 10:772–778.
Bergh T and Lundkvist Ö. (1992) Clinical complications during in-vitro fertilization treatment. Hum Reprod 7:625–626.
Coroleu B, Lopez MF, Hereter L, Veiga A, Calderon G, Martinez F, Carreras O, Barri PN. (1997) Ureteral lesion secondary to vaginal ultrasound follicular puncture for oocyte recovery in in-vitro fertilization. Hum Reprod 12:948–950.
D.I.R. – Deutsches IVF Register 2004 Deutsches IVF Register. (2005).
Dicker D, Ashkenazi J, Feldberg D, Levy T, Dekel A, Ben Rafael Z. (1993) Severe abdominal complications after transvaginal ultrasonographically guided retrieval of oocytes for in vitro fertilization and embryo transfer. Fertil Steril 59:1313–1315.[Web of Science][Medline]
El Shawarby SA, Margara RA, Trew GH, Laffan MA, Lavery SA. (2004) Thrombocythemia and hemoperitoneum after transvaginal oocyte retrieval for in vitro fertilization. Fertil Steril 82:735–737.[CrossRef][Web of Science][Medline]
El-Shawarby SA, Margara RA, Trew GH, Lavery SA. (2004) A review of complications following transvaginal oocyte retrieval for in-vitro fertilization. Hum Fertil 7:127–133.
Fugita OE and Kavoussi L. (2001) Laparoscopic ureteral reimplantation for ureteral lesion secondary to transvaginal ultrasonography for oocyte retrieval. Urology 58:281–282.[Medline]
Govaerts I, Devreker F, Delbaere A, Revelard P, Englert Y. (1998) Short-term medical complications of 1500 oocyte retrievals for in vitro fertilization and embryo transfer. Eur J Obstet Gynecol Reprod Biol 77:239–243.[CrossRef][Web of Science][Medline]
Jones WR, Haines CJ, Matthews CD, Kirby CA. (1989) Traumatic ureteric obstruction secondary to oocyte recovery for in vitro fertilization: a case report. J In Vitro Fert Embryo Transf 6:185–187.[CrossRef][Web of Science][Medline]
Klemetti R, Gissler M, Hemminki E. (2002) Comparison of perinatal health of children born from IVF in Finland in the early and late 1990s. Hum Reprod 17:2192–2198.
Klonoff-Cohen H, Chu E, Natarajan L, Sieber W. (2001) A prospective study of stress among women undergoing in vitro fertilization or gamete intrafallopian transfer. Fertil Steril 76:675–687.[CrossRef][Web of Science][Medline]
Ludwig M, Tölg R, Richardt G, Katus HA, Diedrich K. (1999) Myocardial infarction associated with ovarian hyperstimulation syndrome. JAMA 282:632–633.
Miller P, Price P, Nicholas JE Jr, Hill J. (2002) Acute ureteral obstruction following transvaginal oocyte retrieval for IVF. Hum Reprod 17:137–138.
Roest J, Mous HVH, Zeilmaker GH, Verhoeff A. (1996) The incidence of major clinical complications in a Dutch transport IVF programme. Hum Reprod Update 2:345–353.
Tureck RW, Garcia C, Blaso L, Mastroianni L. (1993) Perioperative complication arising after transvaginal oocyte retrieval. Obstet Gynecol 81:590–593.[Web of Science][Medline]
Wickland M, Lennart M, Hamberger L. (1985) Transvesical and transvaginal approaches for the aspiration of follicles by the use of ultrasound. Ann N Y Acad Sci 442:182–194.[Web of Science][Medline]
Submitted on March 22, 2006; resubmitted on May 2, 2006; accepted on May 5, 2006.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||