Postoperative neuropathy is a well-recognized complication after pelvic surgery. Injury may occur to the nerves coursing through the pelvis or the lower abdominal wall. These complications are often reversible but may be debilitating and result in permanent disability.
We sought to estimate the incidence, mechanism of injury, and outcome of postoperative neuropathies after major pelvic surgery by reviewing our institutional experience over the past 6 years. A secondary purpose was to propose management and prevention schema for postoperative neuropathies.
MATERIALS AND METHODS
The medical records of women who suffered postoperative neuropathy after major pelvic surgery between July 1995 and June 2001 were retrospectively reviewed, and demographic, surgical, postoperative course, and follow-up data were abstracted. These patients were identified by review of the morbidity records from the gynecologic oncology service database at Tampa General Hospital. This database was developed to log events reportable to the American Board of Obstetrics and Gynecology relative to our gynecologic oncology fellowship training program. Patients are identified for entry into the database during their course of care, and information surrounding the event is prospectively recorded. Because these records are used to report our experience to the American Board of Obstetrics and Gynecology, we take special care to see that they are as accurate as possible.
The mechanism of neural injury was determined from the operative report. Diagnosis of neuropathy was confirmed by reviewing the patient history, physical examination findings, consultation reports, and nerve conduction testing when available. Ilioinguinal and iliohypogastric neuropathies were grouped together because of the difficulty in distinguishing individual injuries. Neuropathy duration was defined from the day of surgery until the resolution of symptoms, regardless of when during the postoperative period the diagnosis was made. When symptoms were present at one visit but absent at the subsequent encounter, the latter interval was chosen if the precise time of resolution was not detailed in the patient's record. Symptoms that were present beyond 1 year were defined as persistent.
In efforts to estimate the incidence of neuropathy, the gynecologic oncology surgical procedure database was reviewed to identify the total number of major procedures performed during this same period. Major surgeries were defined as any operative laparoscopy and any procedure through a laparotomy. Minor procedures were not included in the denominator of this proportion because associated neuropathy is very rare after such procedures. The University's Institutional Review Board approved this study.
A total of 1210 women underwent major pelvic surgery on the gynecologic oncology service during the reviewed period. These procedures consisted of 361 hysterectomies for endometrial cancer, 286 ovarian cancer debulkings, 146 radical hysterectomies for cervical cancer, and 417 surgeries for various other benign (eg, endometriosis, adnexal mass), less common malignant (eg, uterine sarcoma), and cancer related (eg, radiation-associated bowel obstruction) conditions. Ninety-one operative laparoscopies were performed, and our retractor of choice for exposure during laparotomy was the Book-walter. Procedures were performed in the supine and lithotomy position in 64% (774) and 36% (436), respectively.
Twenty-four postoperative neuropathies were identified. The median age of these women was 52 years (range 27–80 years), and they were represented by 18 white, three black, one Asian, and two Hispanics. One patient had complete resolution of her symptoms after removal of an epidural that was being used for postoperative analgesia, and she was therefore excluded. The remaining injuries involved the obturator (n = 9), ilioinguinal/iliohypogastric (n = 5), genitofemoral (n = 4), femoral (n = 3), and lumbosacral nerve plexus (n = 2) yielding an incidence of 1.9% (23 of 1210). All but one neuropathy developed in women undergoing surgery for a gynecologic malignancy; the incidences after hysterectomy for endometrial cancer, ovarian cancer debulking, and radical hysterectomy were 1.1% (four of 361), 1.4% (five of 286), and 5.5% (eight of 146), respectively. Tables 1–5 summarize the patients and their respective outcomes stratified by specific neuropathy.
All women with an obturator neuropathy presented with adductor weakness with or without sensory loss over the medial thigh, and most reported symptoms within the first few days after surgery. The mechanism of injury was determined from the operative report, and if no injury was specifically described and the nerve was visualized intact, a stretch injury was presumed.
Obturator nerve resection during exenterative surgery for recurrent cervical cancer led to obturator neuropathy in one patient. The remaining neuropathies were the result of a crush injury by inadvertently including the nerve in a clamp (the nerve was dissected free before suture fixation of the pedicle) while obtaining pelvic sidewall hemostasis in two women, inadvertent division during pelvic lymphadenectomy (immediately repaired) in one patient, and from a presumed stretch injury during lymph node dissection or ovarian tumor debulking in the remaining five subjects. All patients received physiotherapy, and no symptoms persisted beyond 6 months.
The patients who experienced an ilioinguinal or iliohypogastric neuropathy all presented with a lancinating or burning pain in the groin radiating to the vulva. The time to diagnosis varied from immediate to 2 months postoperatively. When pharmacologic management with antiepileptics and/or vitamin B was ineffective, surgical neurolysis was successfully employed. The mechanism of injury was thought secondary to direct nerve injury or entrapment related to the creation or repair of Pfannenstiel incisions in three women, laparoscopic trocar injury in one woman, and was attributed to a stretch injury in a single patient with a midline vertical incision.
Genitofemoral neuropathy was identified in four women who underwent pelvic lymph node dissection. These patients universally presented with “numbness” of the anteromedial thigh from 1–42 days postoperatively. Attempts at symptom management included pyroxidine and gabapentin. Neither of these medications was successful. All injuries resulted in persistent symptoms. The mechanism of injury was presumed to be direct surgical division of the nerve at the time of lymphadenectomy.
Femoral neuropathy was observed in three women. Two patients presented with lower extremity weakness and one with only paresthesia. This latter patient underwent an electromyogram to confirm the diagnosis. All patients received physical therapy and ultimately had complete resolution of their symptoms. All three injuries were attributed to retractor blade compression.
Lumbosacral plexus injury was identified in two women. Efforts to obtain hemostasis in the area of the hypogastric venous plexus during an ovarian debulking resulted in permanent paresthesias of the medial leg and lateral foot along with weak dorsiflexion in one patient. She has had improvement in her symptoms with gabapentin and physical therapy and requires no assistance with ambulation despite persistent deficits. The remaining patient was undergoing a lateral extended endopelvic resection when her lumbosacral plexus was inadvertently divided as it coursed over the piriformis muscle. She suffers permanent disability of the lower leg and requires significant assistance with mobility. She also experiences fecal incontinence, but it is not clear whether this is a complication of her coloanal anastamosis, a result of autonomic denervation, or both.
This report involves a population of patients at increased risk of postoperative morbidity because all women had major surgery, and the nature of disease necessitated radical pelvic dissection for the majority of patients. This high-risk group experienced a 1.9% incidence of postoperative neuropathy, which is consistent with other reports of nerve injury in the general gynecology population.1 We did not include women who underwent minor surgery in our denominator because we did not anticipate (nor find) any neuropathies in that population of patients.
The range in diagnostic timing of postoperative neuropathies is probably related to the underreporting of symptoms or to the lack of attention given to patient complaints by the surgeon. This variability was clearly seen in this series with a wide range in time to diagnosis. Delay did not appear to alter prognosis.
Because of its location, the obturator nerve is most frequently injured during radical or retroperitoneal surgery for gynecologic malignancy or endometriosis.2–4 Obturator neuropathy was the most common injury in the presented patients as it was responsible for 39% (nine of 23) of the observed neuropathies. The diagnosis was made clinically in each of these women. Recovery in the patient who underwent nerve resection was likely a result of compensatory function of other thigh muscles and/or because of the presence of an accessory obturator nerve.5 In general, obturator neuropathy management should include immediate repair of intraoperative transection using microsurgical techniques.3 After repair or postoperative recognition, physiotherapy should be instituted promptly.
Ilioinguinal and iliohypogastric nerve injury were the next most frequently encountered in this series, representing 21.7% (five of 23) of the neuropathies. Injury to these nerves is most commonly seen with lower abdominal transverse incisions, especially those that extend beyond the lateral edge of the rectus into the substance of the internal oblique muscle.6 Symptoms are typically attributed to suture incorporation of the nerve during fascial repair, direct nerve trauma with neuroma formation, or from neural constriction caused by the normal scarring/healing process.7 When patient discomfort is not alleviated with conservative management or nerve block,8 surgical neurolysis, neuroma resection, or nerve resection may be necessary and are generally successful.6,9–11 Ultimately, most patients enjoy full recovery as was demonstrated in this report.
Genitofemoral neuropathy represented 17% (four of 23) of the neuropathies in this series. All patients remain symptomatic, which is probably related to the presumed surgical division of the nerve. We suspect a higher number of patients may suffer this complication but fail to report symptoms given their minor nature.
We observed femoral neuropathy with relative infrequency given the fact that it is the most commonly encountered nerve injury after pelvic surgery with an incidence ranging from 7.45% to 11.6%.12,13 This injury is most often attributed to a retractor compressing the nerve against the pelvic sidewall in the region of the psoas muscle near the inguinal ligament.12,13 Injury may also result from hyperflexion of the thigh by kinking of the nerve under the inguinal ligament when the patient is in the high lithotomy position.14 Femoral nerve injury results in symptoms that vary depending on the degree and location of injury but classically include weakness with hip flexion, leg extension, and sensory loss over the anteromedial thigh and leg, and a diminished or absent knee jerk reflex. With physical therapy, full recovery generally occurs12 as it did in all of our patients. Careful retractor placement and patient positioning are paramount in preventing this injury.12–14
Our observed incidence of 0.17% (two of 1210) for lumbosacral plexus neuropathy is similar to the 0.2% reported by Alsever.1 Both patients who suffered this injury were undergoing very radical resections for malignant disease. Symptoms were reported in the immediate postoperative period, and these were obviously consistent with a nerve trunk injury given the vast distribution of motor and sensory deficits. Both patients improved with physical therapy but neither to the point of complete recovery.
Although common peroneal nerve injury is one of the more common neuropathies encountered in gynecology, we did not identify any patients with such a complication. The two women with lumbosacral injuries did have symptoms in the distribution of the peroneal nerve, but this was attributed to nerve root injury as opposed to a peripheral compression. Likewise, we also did not identify any women with sciatic nerve injury.
Knowledge of the neuroanatomy of the pelvis and lower abdominal wall is necessary for the pelvic surgeon to avoid nerve injury. Important measures for prevention are meticulous attention to surgical technique, care during retractor blade placement, avoidance of extreme lateral extension of lower abdominal transverse incisions, together with avoidance of incorporation of the internal oblique muscle during the fascial repair of these wounds, and particular attention to patient positioning, especially with lithotomy stirrups.
Even when all of these preventative strategies are followed, neuropathy will still occasionally occur. Postoperatively, clinical symptoms and physical examination are usually sufficient to make the diagnosis, and neurologic consultation and testing are not routinely necessary. Physical therapy should be instituted as soon as possible for motor deficits to prevent permanent weakness or atrophy of the involved muscle group. The role of pharmacologic agents such as nutritional supplements (eg, thiamine or pyroxidine), antiepileptics, and tricyclics are unclear. Surgical release or resection may be necessary for women with persistent pain.
Prognosis clearly depends on the nature and severity of the injury; this is supported by the outcomes observed in this series. The only patients with persistent symptoms were those with injury to the lumbosacral plexus and those with an unrepaired transection of the genitofemoral nerve. On the contrary, all patients with a repaired nerve transection, a stretch, or a compression injury had complete resolution of their symptoms within less than 10 months for an overall 73% recovery rate.
1. Alsever JD. Lumbosacral plexopathy after gynecologic surgery: Case report and review of the literature. Am J Obstet Gynecol 1996;174:1769–78.
2. Hoffman MS, Roberts WS, Cavanagh D. Neuropathies associated with radical pelvic surgery for gynecologic cancer. Gynecol Oncol 1988;31:462–6.
3. Vasilev SA. Obturator nerve injury: A review of management options. Gynecol Oncol 1994;53:152–5.
4. Redwine DB, Sharpe DR. Endometriosis of the obturator nerve: A case report. J Reprod Med 1990;35:4–5.
5. Katritsis E, Anagnostopoulow S, Papadopoulos N. Anatomical observations on the accessory obturator nerve (based on 1000 specimens). Anat Anz 1980;148:440–5.
6. Stultz P, Pfeiffer KM. Peripheral nerve injuries resulting from common surgical procedures in the lower portion of the abdomen. Arch Surg 1982;117:324–7.
7. Grosz CR. Iliohypogastric nerve injury. Am J Surg 1981; 142:628.
8. Sippo WC, Burghardt A, Gomez A. Nerve entrapment after Pfannenstiel incision. Am J Obstet Gynecol 1987; 157:420–1.
9. Nahabedian MY, Dellon AL. Outcome of the operative management of nerve injuries in the ilioinguinal region. J Am Coll Surg 1997;184:265–8.
10. Lee CH, Dellon AL. Surgical management of groin pain of neural origin. J Am Coll Surg 2000;191:137–42.
11. Hahn L. Clinical findings and results of operative treatment in ilioinguinal nerve entrapment syndrome. Br J Obstet Gynaecol 1989;96:1080–3.
12. Goldman JA, Feldberg D, Dicker D, Samuel N, Dekel A. Femoral neuropathy subsequent to abdominal hysterectomy. A comparative study. Eur J Obstet Gynecol Reprod Biol 1985;20:385–92.
13. Kvist-Poulsen H, Borel J. Iatrogenic femoral neuropathy subsequent to abdominal hysterectomy: Incidence and prevention. Obstet Gynecol 1982;60:516–20.
14. Dunnihoo DR, Huddleston HT, North S. Femoral nerve palsy as a complication of vaginal hysterectomy: Review of the world literature. J Gynecol Surg 1994;10:1–6.