Postpartum Fever With Pelvic Pain: A Case of Septic Pelvic Thrombophlebitis

Zachary M. Zwolak, DO
Robert C. Langan, MD
Volume 56 - Issue 6 - June 2016

A 20-year-old, gravida 3, para 3 woman presented to the emergency department 11 days postpartum with right lower quadrant abdominal pain that had been present for 3 days. The pain radiated to her right lower back, and an associated fever had been present for 6 days. She described the pain as severe, continuous, and crampy. She complained of associated subjective fevers, chills, decreased appetite, and diarrhea, and she denied having had nausea, vomiting, or dyspnea. She had a scanty, dark vaginal discharge that had been decreasing since delivery.

The patient’s medical history included a precipitous vaginal delivery without instrumentation 11 days prior with an uncomplicated postpartum course. The placenta pathology report was significant for acute, mild chorionitis without chorioamnionitis and mild deciduitis; the patient also had bipolar disorder, gestational anemia, and polycystic ovary syndrome. Her pertinent surgical history included an appendectomy. The patient’s social history was remarkable for having quit smoking 8 months prior; she used no alcohol or illicit drugs and had not been sexually active for the past 3 weeks. Her only medication was ferrous sulfate.

The patient presented in mild distress with a temperature of 38.5°C and a heart rate of 127 beats/min, with moderate right lower quadrant tenderness, and with minimal guarding; the fundus was palpated near the pubic symphysis. The patient could not tolerate a pelvic examination because of pain, but she did describe a scant maroon discharge. The rest of the physical examination findings were normal.

Results of a complete blood cell count were as follows: white blood cell count, 17,690/µL, with 75% segmented neutrophils, 14% lymphocytes, and 10% monocytes; hemoglobin, 10.6 g/dL; hematocrit, 32.1%; and platelet count, 434 × 103/µL.

A computed tomography (CT) scan of abdomen and pelvis demonstrated a mildly distended right ovarian vein with marked surrounding inflammatory changes and low lumen attenuation, suggestive of ovarian thrombophlebitis. 

The patient was admitted and started on a regimen of levofloxacin and metronidazole, along with enoxaparin, 1 mg/kg every 12 hours.  The patient defervesced rapidly, and her symptoms had improved by hospital day 3. The leukocytosis had trended downward, as well. Blood cultures results demonstrated no growth after 5 days.

The patient then decided to leave the hospital against medical advice, and was discharged on a 10-day regimen of levofloxacin and metronidazole, in addition to enoxaparin injections with warfarin therapy with prothrombin time checks for an eventual goal of 2 to 3 seconds (although the current literature does not outline a specific goal for this condition).

Overview OF Septic Pelvic Thrombophlebitis

Septic pelvic thrombophlebitis is divided into 2 categories: ovarian vein thrombophlebitis (OVT) and deep septic pelvic thrombophlebitis, both of which generally present with fever and abdominal pain about 1 week following vaginal or surgical delivery. In 20% of cases, right ovarian vein thrombosis can be seen on CT scans of the abdomen and pelvis.

Septic pelvic thrombophlebitis demonstrates the Virchow triad completely: Peripartum endothelial damage occurs due to delivery or secondary to infection; venous stasis occurs due to pregnancy-induced venous plexus dilation, which frequently leads to retrograde ovarian venous flow from left to right, possibly accounting for the right-sided propensity of this disease; and pregnancy is a relative state of hypercoagulability, especially in the peripartum and postpartum period.

Patients often experience nausea or mild gastrointestinal tract symptoms described as “crampy.”1 Pulmonary emboli occur in 2% of cases but frequently are too small to result in hypoxia.2 Mortality associated with this disease generally is low with intensive medical therapy, and untoward outcomes may result from overwhelming sepsis, although no associated deaths have been reported.2 It is thought that right-sided ovarian vein disease is predominant because of anterograde flow in the right vein and retrograde flow in the left vein based on results of contrast venography.3 The exact etiology of this phenomenon may be due to dextrorotation of the uterus, which compresses the right-sided vasculature.3

This disease occurs in 1 of 9000 vaginal deliveries and 1 of 800 cesarean deliveries.4 A number of factors have been identified that increase the risk for OVT: cesarean delivery, pregnancy, pelvic infection, induced abortion, pelvic surgery, uterine fibroids, and malignancy. The literature suggests that the most significant risk factor is cesarean delivery, which confers a 10-fold risk over vaginal delivery.4

Our patient’s placental pathology suggested a possible infective etiology based on the finding of mild chorionitis and deciduitis.

The diagnosis of OVT is made clinically based on the history, presenting symptoms, leukocytosis, fever, and low abdominal pain; the blood culture results frequently are negative for infection, and imaging studies are not necessarily revealing. Historically, patients who did not defervesce after antibiotic therapy had been started on anticoagulation therapy, although the current literature does not support the use of anticoagulation.4 CT and magnetic resonance imaging are helpful, because thrombus is identified in 20% of cases.4 Ultrasonography is of little utility in the diagnosis of septic pelvic thrombophlebitis.4

The treatment of OVT is medical, although when Von Recklinghausen first described the condition in the 1800s, surgical ligation of the suspect vessel was proposed. Current recommendations are to begin with broad-spectrum antibiotics to cover the likely pathogens associated with endometritis: group A and B streptococcus, coliform bacteria, and anaerobes. The length of treatment is debatable, varying from 48 hours if leukocytosis normalizes, to a 7-day course.

Experts previously had recommended either anticoagulation with heparin, 5000 U bolus, followed with 16 to 18 U/kg, for a goal partial thromboplastin time of 1.5 to 2 times the patient’s baseline, or anticoagulation with enoxaparin, 1 mg/kg every 12 hours for a duration of 2 weeks up to 6 weeks, but comparative studies are lacking.1,5 There is consensus that in cases of an underlying hypercoagulable state, therapy should be continued longer. 

Despite the clinical sense of anticoagulation, the results of a randomized trial conducted by Brown and colleagues4 refutes the use of anticoagulants, with the duration of fever being similar in the group given heparin and antimicrobials compared with the group treated solely with antimicrobial agents.

Zachary M. Zwolak, DO, is on the family medicine faculty at the Ventura County Medical Center Family Medicine Residency Program in Ventura, California.

Robert Langan, MD, is the director of the St. Luke’s Family Medicine Residency Program in Bethlehem, Pennsylvania.

References:

  1. Witlin AG, Sibai, BM. Postpartum ovarian vein thrombosis after vaginal delivery: a report of 11 cases. Obstet Gynecol. 1995;85(5 pt 1):775-780.
  2. Chen KT. Septic pelvic thrombophlebitis. UpToDate. http://www.uptodate.com/contents/septic-pelvic-thrombophlebitis. Accessed May 3, 2016.
  3. Klima DA, Snyder TE. Postpartum ovarian vein thrombosis. Obstet Gynecol. 2008;111(2 pt 1):431-435.
  4. Brown CE, Stettler RW, Twickler D, Cunningham FG. Puerperal septic pelvic thrombophlebitis: incidence and response to heparin therapy. Am J Obset Gynecol. 1999;181(1): 143-148.
  5. Tang LCH, Woo JSK, Choo YC. Puerperal ovarian vein thrombophlebitis. Postgrad Med J. 5;61(712):179-180.