Sonographic findings in acute puerperal endometritis

Abstract

Purpose

To evaluate the reliability of the sonographic subserosal hypoechoic rim sign and endomyometrial junction indistinctness in distinguishing patients with acute puerperal endometritis from other common postpartum complications, particularly those with retained products of conception.

Methods

Radiographic coding identified all patients presenting to Auckland City Hospital over a 12‐month period between 1 March 2016 and 28 February 2017 who had pelvic ultrasound scans to assess for postpartum complications. Clinical notes were reviewed to identify those patients with a clinical diagnosis of acute endometritis. After exclusion criteria were applied, the ultrasound images of 31 patients with acute endometritis and 31 randomly selected controls from the remaining pool of postpartum patients were randomised and anonymised. Ultrasound images were retrospectively reviewed by two independent radiologists to identify the presence or absence of the hypoechoic subserosal rim and endomyometrial junction indistinctness signs.

Results

The average sensitivity and specificity for readers identifying the subserosal hypoechoic rim sign in full‐term postpartum patients with acute endometritis were 76.9% and 92.1%, respectively. Interobserver reliability was moderate with a kappa value of 0.5. The average sensitivity and specificity for endomyometrial junction indistinctness were 89.0% and 95.2%, respectively, with substantial inter‐observer agreement, Kappa 0.7. These signs remained specific but were less frequently observed in postpartum patients scanned during early pregnancy.

Conclusion

We have shown that identifying the subserosal hypoechoic rim and endomyometrial junction indistinctness on ultrasound scans can reliably help to distinguish patients with acute postpartum endometritis from other complications, particularly in full‐term postpartum patients.

Introduction

Acute postpartum endometritis complicates 2–3% of vaginal deliveries and up to 28% of caesarean sections. Use of antibiotic prophylaxis reportedly reduces this number but up to 11% of caesarean section deliveries develop endometritis despite this precaution.1

Puerperal sepsis is among the leading causes of maternal mortality worldwide, accounting for one tenth of the global burden of maternal deaths.2 Attempts to prevent postpartum infection have included reducing common risk factors such as prolonged rupture of membranes, prolonged labour, multiple vaginal examinations, and the use of prophylactic antibiotics.2, 3 However, early diagnosis and prompt treatment remain imperative for reducing short‐ and long‐term complications of acute endometritis.3

Endometritis is the most common cause of fever in the puerperal period, and the diagnosis is considered to be primarily clinical. However, clinical findings suggestive of this diagnosis such as uterine tenderness and leucocytosis are non‐specific and can also be observed in normal postpartum patients.1, 4 The sonographic appearance of endometritis traditionally has also been reported as non‐specific because of significant overlap between the physiologic appearance and dimensions of the normal postpartum uterus and those patients with endometritis.4

We recently described the sonographic appearance of patients with clinical endometritis and showed that the subserosal hypoechoic rim (SHR) and endomyometrial junction indistinctness (EMI) reliably distinguished those patients with a clinical diagnosis of acute endometritis from normal controls.5

The aim of this study was to determine if these signs could help to distinguish patients with acute endometritis from other common puerperal complications. Prolonged bleeding is one of the more common symptoms for which postpartum ultrasound (US) is requested, and these scans are performed most frequently to assess for the presence of retained products of conception (RPOC) among other complications.6 Sonographic findings of focal echogenic endometrial thickening with increased colour Doppler are typical for RPOC but these findings can also be seen to a variable degree in the normal postpartum uterus. High false‐positive rates for identifying RPOC have been reported when US findings were correlated with histology.4 Differentiation between complications such as RPOC and endometritis is critically important as RPOC benefits from intervention, whereas intervention in patients with endometritis is known to increase complications.6

Materials and Methods

This was a retrospective study that included all patients in whom transvaginal scans were performed at Auckland City Hospital to assess for postpartum complications during the 12 months between 1 March 2016 and 28 February 2017. These included patients with suspected acute endometritis and RPOC (n = 436). Routine transvaginal US images were performed by qualified sonographers using either a Phillips IU22, Philips EPIQ 7G or GE V830 machines.

Electronic clinical notes and discharge summaries were retrospectively reviewed to identify patients in whom a clinical diagnosis of acute endometritis had been made. Clinical data including inflammatory biomarkers within 3 days of US, duration of pregnancy and exposure to antibiotics prior to US were also recorded.

A total of 62 patients met the eligibility criteria for this study. All patients were scanned postpartum following vaginal deliveries, caesarean sections, as well as after miscarriage, termination of pregnancy, and evacuation of retained products of conception (ERPOC).

Patients were excluded if they were scanned outside the puerperal period, defined as 6 weeks postpartum,7 or if they were found to have another cause of postpartum fever. 31 patients with acute endometritis met the inclusion criteria, fourteen of whom had concurrent RPOC. Control patients were randomly selected on a 1:1 basis from the remaining pool of patients in whom postpartum ultrasound scans had been performed.

All transvaginal US images were anonymised and randomised, then retrospectively reviewed by two consultant radiologists independently, both of whom have a special interest in gynaecological imaging. Both reviewers were involved in the previous study4 and were familiar with the appearance of the SHR and EMI signs. No US images of patients from the previous study were included in this study, and the reviewers were blinded to all patient information. Each reader independently recorded the presence or absence of the SHR and EMI signs on the available US images for each case.

Results were analysed to determine the sensitivity and specificity, positive and negative predictive values, and positive and negative likelihood ratios of each sign for each reader. 95% confidence intervals were calculated using the exact Clopper‐Pearson confidence interval and the log method from Altman et al. (2000). Inter‐reader concordance was also calculated, and Kappa values were used to establish the strength of inter‐observer agreement for each sign. Subgroup analysis was performed in full‐term postpartum patients and in early pregnancy patients.

Ethics approval was obtained through the ADHB Research Office with waiver for informed consent (A + 7194).

Results

Of the 62 patients included in this study, 31 had a clinical diagnosis of acute endometritis, 14 of whom had concurrent RPOC. Of the 31 randomly selected control cases, 23 had RPOC and 8 had been interpreted as normal. Table 1 outlines the mode of delivery for both groups. 29% of the endometritis cases occurred after caesarean section. 77% of the control group were imaged following miscarriage.

Table 1.

Mode of delivery

Mode of delivery	Clinical endometritis cases (n = 31)	Controls (n = 31)
Term caesarean section	9 (29%)	1 (3%)
Term vaginal delivery	5 (16%)	0 (0%)
Termination of pregnancy	8 (26%)	3 (10%)
Miscarriage	4 (13%)	24 (77%)
Evacuation of retained products of conception	5 (16%)	3 (10%)

The average duration of pregnancy for the 17 patients in the endometritis group who were not full term was 10.2 weeks with a range of 6–18 weeks. Except for the patient who was delivered by caesarean section, the average duration of pregnancy for the control group was 9.8 weeks with a range of 5–16 weeks.

Table 2 outlines the demographic, inflammatory biomarker and antibiotic treatment data for each group. The average ages for the study and control patients were 30 and 31 years, respectively. There was no significant difference in white cell count (WCC) between the two groups. C‐reactive protein (CRP) was measured in 23 endometritis patients, of which 14 (60.9%) were raised with an average value of 59. Only one control patient had CRP measured, which was normal. 80.6% of the endometritis patients received antibiotics prior to US, for an average duration of 2.3 days. None of the control patients had antibiotics prior to US.

Table 2.

Patient demographics, inflammatory biomarkers and duration of antibiotics

Demographic/biochemistry	Clinical endometritis cases (n = 31)	Controls (n = 31)
Mean Age (years)	30	31
Documented fever (>38°C)	14 (45.2%)	0
Raised WCC	7 (22.6%)	6 (25.0%)a
Average WCC	9.6 (range 4.4–30.6)	8.9 (range 4.5–15.2)b
Raised CRP	14 (60.9%)b	0c
Average CRP	59 (range <0.6–514)	1.6c
Received oral or IV antibiotics for >1 day prior to USS	25 (80.6%)	0
Average duration of IV or oral antibiotics prior to USS (days)	2.3 (range <1–12)	0

^a7 of the control group had no WCC measured.

^b8 of the endometritis group had no CRP measured.

^cOnly one control patient had a CRP measured.

Subserosal Hypoechoic Rim (SHR) sign

Sensitivity for identifying the SHR sign in all patients with acute endometritis was 58.1%, 95% CI [39.1, 75.5] for Reader 1, and 46.7%, 95% CI [28.3, 65.7] for Reader 2 (Table 3). When the presence of the SHR sign was assessed in the subgroup of full‐term delivery patients, the sensitivity increased to 84.6%, 95% CI [54.6, 98.1] for Reader 1 and 69.2%, 95% CI [38.6, 90.9] for Reader 2. When the results for Readers 1 and 2 were concordant, the sensitivity for term delivery patients (including vaginal delivery and caesarean section) was 81.8%, 95% CI [48.2, 97.7].

Table 3.

Results subserosal hypoechoic rim sign

Reader	Overall sensitivity	Full‐term postpartum sensitivity	Early pregnancy postpartum sensitivity	Overall specificity	Positive likelihood ratio	Negative likelihood ratio
1	58.1% 95% CI [39.1, 75.5]	84.6% 95% CI [54.6, 98.1]	38.9% 95% CI [17.3, 64.3]	93.6% 95% CI [78.6, 99.2]	9.0 95% CI [2.3, 35.5]	0.5 95% CI [0.3, 0.7]
2	46.7% 95% CI [28.3, 65.7]	69.2% 95% CI [38.6, 90.9]	27.8% 95% CI [9.7, 53.5]	90.6% 95% CI [75.0, 98.0]	4.98 95% CI [1.6, 15.6]	0.6 95% CI [0.4, 0.8]
Average	52.4% ± 8.1	76.9% ± 10.9	33.4% ± 7.8	92.1% ± 2.1	7.0 ± 2.8	0.5 ± 0.1
1 & 2	52.2% 95% CI [30.6, 73.2]	81.8% 95% CI [48.2, 97.7]	25.0% 95% CI [5.5, 57.2]	100% 95% CI [86.8, 100]	a	0.5 95% CI [0.3, 0.7]

^aPLR unable to be calculated as specificity = 100%.

Specificity of the SHR sign in all patients was 93.6%, 95% CI [78.6, 99.2] for Reader 1, and 90.6%, 95% CI [75.0, 98.0] for Reader 2. When concordant between readers, the overall specificity of the SHR sign was 100%, 95% CI [86.8, 100].

The positive likelihood ratios for each reader showed moderate increase in the probability of endometritis when the SHR sign was present: 9.0, 95% CI [2.3, 35.5] for Reader 1, and 4.98, 95% CI [1.6, 15.6] for Reader 2.

Concordance between the two readers for the SHR sign was 79.0%. The inter‐observer agreement was moderate with a Kappa value of 0.5.

Endomyometrial Junction Indistinctness (EMI) sign

Sensitivity of the EMI sign in all patients with acute endometritis was 54.8%, 95% CI [36.0, 72.7] for Reader 1, and 58.1%, 95% CI [39.1, 75.5] for Reader 2 (Table 3). When the presence of EMI was assessed in the subgroup of patients following full‐term delivery, the sensitivity increased to 92.2%, 95% CI [64.0, 99.8] for Reader 1, and 85.7%, 95% CI [57.2, 98.2] for Reader 2. When the results of Readers 1 and 2 were concordant, the sensitivity for this group was 91.7%, 95% CI [61.5, 99.8] (Table 4).

Table 4.

Results endomyometrial junction indistinctness sign

Reader	Overall sensitivity	Full‐term postpartum sensitivity	Early pregnancy postpartum sensitivity	Overall specificity	Positive likelihood ratio	Negative likelihood ratio
1	54.8% 95% CI [36.0, 72.7]	92.2% 95% CI [64.0, 99.8]	27.8% 95% CI [9.7, 53.5]	96.8% 95% CI [83.3, 99.9]	17.0 95% CI [2.4, 120.0]	0.5 95% CI [0.3, 0.7]
2	58.1% 95% CI [39.1, 75.5]	85.7% 95% CI [57.2, 98.2]	38.9% 95% CI [17.3, 64.3]	93.6% 95% CI [78.6, 99.2]	9.0 95% CI [2.3, 35.5]	0.5 95% CI [0.3, 0.6]
Average	56.5% ± 2.3	89.0% ± 4.6	33.4% ± 7.8	95.2% ± 2.3	13.0 ± 5.7	0.5 ± 0
1 & 2	58.3% 95% CI [36.6, 77.9]	91.7% 95% CI [61.5, 99.8]	25.0% 95% CI [5.5, 57.2]	96.7% 95% CI [82.8, 99.9]	17.5 95% CI [2.5, 123.8]	0.4 95% CI [0.3, 0.7]

Specificity for the EMI sign in all patients was 96.8%, 95% CI [83.3, 99.9] for Reader 1, and 93.6%, 95% CI [78.6, 99.2] for Reader 2. When concordant between readers, the overall specificity for the EMI sign was 96.7%, 95% CI [82.8, 99.9].

The positive likelihood ratio for Reader 1 was 17.0, 95% CI [2.4, 120.0], indicating a large increase in the probability of endometritis when EMI was identified. The positive likelihood ratio for Reader 2 was moderate at 9.0, 95% CI [2.3, 35.5].

Concordance between the two readers for the EMI sign was 87.1%. Inter‐observer agreement was substantial with a Kappa of 0.7.

Subserosal hypoechoic rim (SHR) and endomyometrial junction indistinctness (EMI) signs

The concordant overall sensitivity for the presence of both signs was 58.8%, 95% CI [32.9, 81.6], and overall specificity was 100%, 95 CI [86.3, 100].

Concordant sensitivity for term postpartum patients only was 88.9%, 95% CI [51.8, 99.7] (Table 5).

Table 5.

Results subserosal hypoechoic rim and endomyometrial junction indistinctness signs when both present or absent

Reader	Overall sensitivity	Full‐term postpartum sensitivity	Early pregnancy postpartum sensitivity	Overall specificity	Positive likelihood ratio	Negative likelihood ratio
1	57.1% 95% CI [37.2, 75.5]	91.7% 95% CI [61.5, 99.8]	31.3% 95% CI [11.0, 58.7]	96.7% 95% CI [82.8, 99.9]	17.1 95% CI [2.4, 120.9]	0.4 95% CI [0.3, 0.7]
2	57.1% 95% CI [34.1, 78.2]	88.9% 95% CI [51.8, 99.7]	27.3% 95% CI [6.0, 61.0]	100% 95% CI [86.8, 100]	a	0.4 95% CI [0.3, 0.7]
Average	57.1% ± 0	90.3% ± 2.0	29.3% ± 2.8	98.4% ± 2.3	a	0.4 ± 0
1 & 2	58.8% 95% CI [32.9, 81.6]	88.9% 95% CI [51.8, 99.7]	25.0% 95% CI [3.2, 65.1]	100% 95% CI [86.3, 100]	a	0.4 95% CI [0.2, 0.7]

^aPLR unable to be calculated as specificity = 100%.

Subgroup analysis of those patients both from the endometritis group and groups who were scanned following early pregnancy retained the high specificity for these signs but the average sensitivity decreased to 33.4% for each of the SHR and EMI signs (Figure 1).

Figure 1.

(a) Sagittal Transvaginal Ultrasound of the Uterus in a 23‐Year‐Old Patient 23 days Post‐caesarean Section Delivery with a Clinical Diagnosis of Postpartum Endometritis Showing Both the Subserosal Hypoechoic Rim Sign and Indistinctness of the Endomyometrial Junction. (b) Sagittal Transvaginal Ultrasound of the Uterus in an 18‐Year‐Old Patient 11 days Post‐miscarriage at 10 weeks’ Gestation with a Clinical Diagnosis of Endometritis Showing Both the Subserosal Hypoechoic Rim Sign and Indistinctness of the Endomyometrial Junction with Concurrent Retained Products of Conception.

Discussion

Acute postpartum endometritis remains the most common cause of puerperal sepsis and a major cause of maternal mortality.2 Women are at risk of endometritis after all methods of fetal delivery including miscarriage, termination of pregnancy as well as delivery of term babies. However, caesarean section carries the greatest risk. Even with antibiotic prophylaxis, approximately 11% of caesarean section deliveries will be complicated by acute puerperal endometritis.1, 8 Early diagnosis and treatment remain the keys to reducing short‐ and long‐term complications, which include pelvic inflammatory disease, infertility and death.

The World Health Organization (WHO) recommends treating patients with postpartum endometritis with a combination of clindamycin and gentamicin for at least 24–48 h after complete resolution of clinical signs and symptoms.2 However, diagnosis remains a challenge and generally is based on clinical criteria that include fever, vaginal discharge and uterine tenderness. Both clinical and ultrasound findings in postpartum endometritis are described as non‐specific, frequently resulting in empiric antibiotic treatment of patients.4 More accurate diagnosis of endometritis will not only help to expedite appropriate treatment but should reduce excessive use of antibiotics. Postpartum antibiotics can not only cause adverse reactions to mother and baby, but also increase the risk of antimicrobial resistance with wider public health implications.

Previous attempts to correlate sonographic findings with the clinical diagnosis of endometritis include measuring the AP diameter of the uterus and uterine cavity by Mulic‐Lutvica et al. They found no significant difference in these variables or other uterine features between postpartum patients with endometritis and those without complications in the puerperal period.9

Gillies et al. recently described the sonographic hypoechoic rim sign (SHR) and endomyometrial junction indistinctness (EMI) in patients with acute endometritis.5 They compared these findings in patients with clinically diagnosed endometritis with a control group of women scanned for other reasons, some of whom were postpartum but most of whom had not been pregnant. They found that these two sonographic features reliably identified the patients with endometritis. Our challenge was to assess the reliability of these signs in distinguishing patients with clinical endometritis from other patients with a variety of postpartum problems for which women are frequently scanned. For this reason, our control group was randomly selected from patients who had been scanned to assess for postpartum complications but whose scans had been interpreted as either RPOC or normal.

Our results support the findings of Gillies et al. and show that the SHR and EMI signs can help to identify endometritis, even in those patients who have both RPOC and endometritis. We found both signs to be highly specific and therefore reliable for identifying patients who were not infected. This ability to rule out the diagnosis of endometritis effectively will not only help to prevent women from unnecessary antibiotics but can direct their physicians to the appropriate investigation and therapy of other sources of infection indicated by the presence of fever or raised inflammatory markers. Gram‐negative sensitive antibiotics used for suspected endometritis are very different from those used for pneumonia, for example.

The sensitivity of both signs increased when considering only the subgroup of patients after full‐term delivery which we found interesting. This could be explained by the physiological changes that occur in the pregnant uterus, particularly in the third trimester as well as during involution of the uterus that occurs in the acute puerperal period.10 Compared with the first trimester, a term uterus demonstrates increased vascularity. This likely facilitates the increase in permeability and oedema as a response to infection. The role of lymphatics in the uterus is poorly understood but may contribute to the sonographic signs we have observed.

As in other organs, increased lymphangiogenesis occurs within the pregnant uterus as a complement to increased vascularity. In analysing changing lymphatic function in the pregnant uterus, Red‐Horse postulates that decidual lymphatic vessels could be important as regulators of fluid homeostasis and that the increased blood flow to the uterus, as the patient nears term, results in increased interstitial fluid that must be cleared by uterine lymphatics.11 It seems a reasonable extrapolation that the increased blood flow due to infection will also increase capillary leakage and result in accumulation of oedema/extracellular fluid, ultimately to be removed by the lymphatics. When oedema increases in the infected uterus, however, engorged lymphatics could give rise to both the subserosal rim and endomyometrial junction indistinctness. The relatively rigid uterine serosa would act as a barrier against which the engorged lymphatics and interstitial fluid could accumulate to create the hypoechoic rim we have described.

It is probable that indistinctness of the endomyometrial junction is also due to oedema that spreads from the endometrium through to the myometrium. The subserosal hypoechoic rim also likely reflects oedema when inflammation has permeated the full thickness of the myometrium. This could explain why we observed endomyometrial indistinctness more frequently than the subserosal hypoechoic rim with a sensitivity of 91.7%, compared with 81.8%. Oedematous blurring of the myoendometrial junction likely precedes full thickness oedema that extends to the serosal surface. Myoendometritis is perhaps a more appropriate term than endometritis as it describes the migratory inflammatory process within the uterus.

We found the decreased sensitivity of both signs in patients who were scanned following early pregnancy loss to be surprising but important.

These patients had been scanned mostly in the first trimester after miscarriage, termination or evacuation of retained products. Subgroup analysis of these patients (endometritis n = 17, controls n = 30) showed a considerable decrease in sensitivity for both readers but their specificities were maintained. In patients scanned during their first trimester, absence of the hypoechoic rim and EMI did not exclude the diagnosis of endometritis as reliably as in full‐term patients.

The false‐positive and false‐negative data are taken from those results that were concordant between Readers 1 and 2.

There were no false‐positive SHR cases. The only false‐positive EMI sign was in a postpartum patient after full‐term delivery who also had a large 13.8 cm fibroid distorting the uterine cavity.

Of the false‐negative SHR cases, 8 of 10 (80%) were seen in patients scanned in their first trimester of pregnancy. Only 2 followed full‐term delivery, one of which had a 6.8‐cm submucosal fibroid (Figure 2a), and the other required surgical intervention for concurrent RPOC. Her pathology was described as having ‘low grade endometritis’ in the surgical operation note. Three false‐negative SHR cases in whom the submucosal hypoechoic rim was not seen had histopathological specimens, none of which showed acute endometritis although one showed chronic endometritis. Because these patients were treated clinically as acute endometritis, they were included as a false negatives. The false‐negative SHR patients had similar inflammatory biomarkers and duration of antibiotic therapy prior to US as the SHR patients who were true positives.

Figure 2.

(a) Sagittal Transvaginal Ultrasound of the Uterus in a 36‐Year‐old Patient 22 days Post‐caesarean Section with Postpartum Endometritis and a Large Fibroid Read as Negative for Subserosal Hypoechoic Rim and Endomyometrial Junction Indistinctness. (b) Sagittal Transvaginal Ultrasound of the Uterus in a 31‐Year‐old Patient 4 days Post‐miscarriage at 7 weeks’ Gestation Without Endometritis Showing RPOC Only, with Preservation of the Distinctness of the Endomyometrial Junction Elsewhere and Lack of a Subserosal Hypoechoic Rim.

Of the false‐negative EMI findings, 8 of 9 (89%) scans were performed after miscarriage, ERPOC or termination following early pregnancy. The one patient who developed clinical endometritis after full‐term delivery had a large submucosal fibroid distorting the endometrium and the hypoechoic rim was also not identified by either reader on her scans (Figure 2a). Four of these patients had histopathological specimens, none of which showed acute endometritis. However, two demonstrated chronic endometritis but were treated as acute endometritis so were included as false negatives. The false‐negative EMI cases had similar inflammatory biomarkers and duration of antibiotics prior to scanning as the true positive cases.

Correlation between the clinical and ultrasound diagnoses of acute endometritis with histopathology is challenging because only those patients with concurrent RPOC are likely to require intervention with availability of pathological specimens. Acute endometritis is diagnosed histologically by the presence of clusters of neutrophils, whereas the presence of plasma cells is pathognomonic of chronic endometritis.12 Therefore, acute endometritis is difficult to diagnose in an RPOC specimen when reactive neutrophils are expected in the endometrium underlying retained products. To reliably diagnose acute endometritis histologically, separate endometrial samples are required. This may explain why no patients who were treated for both acute endometritis and retained products that required intervention were reported to have acute endometritis on their histopathological specimens and chronic endometritis was reported in only two because of the presence of plasma cells.

A limitation of our study is the predominance of first trimester patients in the randomly selected control group. However, it has only been through this analysis, that we have learned of the lower sensitivity of these sonographic signs in patients scanned following early pregnancy loss.

A second limitation that the readers found was variation in image quality that was due to technical and operator differences. We believe that awareness of the significance of these two signs will result in more attention to the importance of acquiring standard midline transvaginal sagittal images that include the whole uterus as well as technical aspects such as time‐gain control for achieving appropriate levels of endomyometrial contrast. Both the endomyometrial junction and subserosal rim can be assessed most effectively with a sagittal view that includes the whole uterus. We found the subserosal surface of the uterus to be more susceptible to artefact, particularly if the focus was not appropriate and if the image contrast was too high.

A further limitation is the relatively small sample size of our study. Despite reviewing a large pool of patients (436), the number of patients with the clinical diagnosis of acute endometritis was relatively modest (n = 31). These numbers reduced further when broken down for subgroup analyses of full‐term and early pregnancy postpartum patients (14). Despite this, our overall results are consistent with the findings of Gillies et al.5 and show that identifying the SHR and EMI signs on US scans can help to identify patients with endometritis, even in patients with concurrent RPOC.

Conclusion

We have shown that the subserosal hypoechoic rim (SHR) sign and indistinctness of the endomyometrial junction (EMI) are useful in diagnosing acute myoendometritis, particularly in full‐term postpartum patients while demonstrating strong specificity amongst those patients with the other most common postpartum complication, retained products of conception.

Disclosure statement

No conflicts of interests to declare.

Authorship declaration

This authorship listing conforms with the journal's authorship policy, and all authors are in agreement with the content of the submitted manuscript.