A 19‐Year Audit of Genital Mycoplasma Species in Blood Cultures From Women Attending a Tertiary Maternity Hospital

ABSTRACT

Background

The genital mycoplasmas, Ureaplasma parvum, Ureaplasma urealyticum and Mycoplasma hominis commonly colonise the female genital tract and occasionally cause invasive infection. Our microbiology laboratory, which services a specialist maternity referral hospital, is unique in manufacturing specialised in‐house blood culture media for detecting mycoplasma bacteraemia. This audit aimed to review the proportion of patients with a positive blood culture for a genital mycoplasma species over a 19‐year period and to describe clinical management and outcomes for these patients.

Methods

Data was extracted from the laboratory information system to identify all mycoplasma blood culture bottles collected from 2000 to 2018 for women > 18 years at the Royal Women's Hospital in Victoria, Australia. Clinical and demographic data for women with a blood culture positive for a genital mycoplasma species over the period 2008–2018 were extracted from the hospital medical records.

Results

The average annual blood culture positivity rate for a genital mycoplasma was 0.56%. Thirty‐eight women had a blood culture positive for Ureaplasma spp. (n = 28) or Mycoplasma hominis (n = 13); three women had both organisms isolated from a single blood culture. Most women had a genital mycoplasma isolated in the early post‐partum period (66%) or associated with early pregnancy loss (23%). 57% of women received antibiotics which would be expected to have activity against mycoplasmas. No adverse outcomes were recorded within 30 days, regardless of whether mycoplasma‐active antibiotics were prescribed.

Conclusion

Genital mycoplasma bacteraemia may occur in the early post‐partum period or be associated with early pregnancy loss. It is unclear whether this represents transient bacteraemia or whether women benefit from directed antimicrobial therapy.

1. Background

Ureaplasma parvum, Ureaplasma urealyticum and Mycoplasma hominis are collectively known as the genital mycoplasmas. Colonisation with genital mycoplasmas is very common [1]; up to 80% of sexually active women will have Ureaplasma spp. detected from the genital tract while up to 50% will be colonised with M. hominis [2]. These genital mycoplasmas are usually regarded as normal microbial flora, unlike Mycoplasma genitalium , which is considered more pathogenic and a cause of sexually transmissible infection [3].

The presence of the genital mycoplasma organisms in the female genital tract has been associated with a variety of conditions including bacterial vaginosis, pelvic inflammatory disease, infertility and spontaneous preterm labour; however, their significance as pathogens in many disease states remains uncertain [2, 4, 5, 6, 7, 8]. There is controversy over the utility of screening for genital mycoplasmas in predefined populations, such as pregnant women [5]. In published studies, detection of ureaplasma in amniotic fluid has been associated with preterm labour [2, 9, 10, 11]. It remains unclear what patient factors lead to ascending or disseminated infection with genital mycoplasmas, although disruption of the genital tract flora is commonly cited [12].

Bacteraemia with genital mycoplasmas has been reported in post‐partum women [6, 13, 14]. The significance of bacteraemia with a genital mycoplasma remains to be fully elucidated. Isolation of a genital mycoplasma from blood may prompt treatment by clinicians. It is not known, however, whether treatment is necessary to optimize patient outcomes, although response to therapy has been demonstrated in animal models [15].

Routine laboratory microbiology will not detect the genital mycoplasmas, as they are fastidious bacteria that lack a cell wall and require specialised culture media for detection. The Royal Children's Hospital (RCH) microbiology laboratory, which services the Royal Women's hospital (RWH), a tertiary maternity hospital, provides access to specialised blood culture media for detecting bacteraemia with genital mycoplasmas. Mycoplasma culture bottles are capable of detecting Ureaplasma spp. and M. hominis , but not M. genitalium . Mycoplasma blood culture bottles are manufactured in‐house, which is not routinely done by most hospital microbiology laboratories.

Over the past three decades, our hospital guidelines have recommended collection of a mycoplasma blood culture bottle in the setting of suspected peripartum sepsis and postpartum fever, in addition to standard aerobic/anaerobic blood culture bottles. If clinicians feel that it is clinically appropriate, they may choose to collect a mycoplasma blood culture for other indications such as fever following termination of pregnancy or a gynaecological procedure.

As our laboratory is unique in having mycoplasma blood culture media available for routine diagnostic use since the 1980s, this study represents one of the larger data sets providing information about bacteraemia with genital mycoplasmas in a tertiary maternity hospital.

The aims of this audit are to review the proportion of patients from our hospital with a positive blood culture for a genital mycoplasma species over a 19‐year period, and to describe the clinical management and outcomes for these patients.

2. Methods

2.1. Laboratory Testing

Over the study period, mycoplasma blood culture bottles were produced by the RCH laboratory and supplied to the RWH. Mycoplasma blood culture bottles are manufactured in‐house using a ureaplasma broth (modified A8 media), a tryptone soya broth (tryptone soya broth (Oxoid), NaCl and KH₂PO₄) with additives (10% urea, horse serum, phenol red, ampicillin, and vancomycin). The mycoplasma blood culture bottles were made available on clinical wards at the RWH. They were supplied bundled with standard aerobic and anaerobic blood culture bottles (BacT/ALERT, Biomerieux, Marcy‐l'Etoile, France).

Throughout the study period, although it was at the discretion of the clinician, hospital guidelines recommended collection of a mycoplasma blood culture bottle in the setting of suspected peripartum sepsis. Mycoplasma bottles were aseptically inoculated with 3–5 mL of blood using a syringe.

Blood culture bottles received by the laboratory were incubated at 35°C for 48 h. After 48 h, samples were inoculated onto ureaplasma A8 agar using a sterile loop. Plates were incubated anaerobically at 35°C for 48 h, after which colonies were identified morphologically using a plate microscope.

Some genital mycoplasma culture isolates were further speciated as Ureaplasma parvum or Ureaplasma urealyticum using a validated in‐house PCR, which has been previously described in detail [16]. Briefly, DNA from colonies was extracted using an automated MagNA Pure 96 extraction system (Roche Diagnostics, Mannheim, Germany). Real‐time PCR, using Taqman probes for Ureaplasma parvum (5′‐TCCACAAGCTCCAGCAGCAATTTG‐3′) and Ureaplasma urealyticum (5'ACCACAAGCACCTGCTACGATTTGTTC‐3′), was performed on a LightCycler 480 (Roche Diagnostics). We did not use molecular methods for further identification of Mycoplasma hominis.

BacT/ALERT bottles for standard aerobic and anaerobic culture were incubated for 5 days in the BacT/Alert 3D system (Biomerieux, Marcy‐l'Etoile), and sub‐cultured onto horse blood agar and chocolate agar if they flagged positive, with further biochemical identification to species level.

2.2. Data Extraction

Data was extracted from the RCH laboratory information system (Medipath) to identify all mycoplasma blood culture bottles collected from women over 18 years of age at the RWH, from 2000 to 2018 inclusive. The proportion of positive blood cultures for genital mycoplasmas was calculated by year.

2.3. Clinical Data

Medical records were available for patients with blood cultures collected after 2008. Where available, the following information was obtained for patients with a positive blood culture for a genital mycoplasma: patient age, comorbidities including current or past pregnancy as well as any relevant complications; clinical details: presenting symptoms, antibiotics administered, and patient outcome. Positive results for genital mycoplasmas from sites other than blood were also recorded.

2.4. Ethics Statement

The study was endorsed by the RWH Research Committee and RWH Human Research Ethics Committee, who confirmed that it meets the National Health and Medical Research Council requirements for quality assurance/audit projects, AQA20/10.

3. Results

3.1. Laboratory

Over the study period (2000–2018), there were 6 669 Mycoplasma blood culture bottles received, of which 41 were positive from 38 women, with an average positivity rate of 0.56%. There were minor variations in the culture positivity rate over the 20‐year period (see Table 1); the highest number of positive cultures occurred in 2013 (n = 7; positivity rate 1.43%).

TABLE 1.

Annual mycoplasma blood culture positivity over a 19‐year period.

Year	Positive bottles	Positive patients	Mycoplasma blood cultures collected	% positivity (by patient)
2000	2	2	212	0.94
2001	0	0	224	0
2002	0	0	281	0
2003	1	1	297	0.34
2004	0	0	305	0
2005	3	3	373	0.80
2006	1	1	354	0.28
2007	2	1	370	0.27
2008	3	2	367	0.54
2009	2	2	306	0.65
2010	1	1	282	0.35
2011	4	4	304	1.31
2012	2	2	470	0.43
2013	7	7	491	1.43
2014	4	4	486	0.82
2015	2	2	466	0.43
2016	2	2	409	0.49
2017	3	3	395	0.76
2018	2	1	277	0.36
	41	38	6669	0.56

Of the 38 women identified to have a positive blood culture with a genital mycoplasma, 28 had a Ureaplasma spp. isolated and 13 had Mycoplasma hominis isolated. There were three patients with both Ureaplasma spp. and Mycoplasma hominis isolated from the same bottle.

3.2. Clinical Characteristics

From 2008 to 2018, there were 30 women with blood cultures positive for Ureaplasma urealyticum, Ureaplasma parvum or Mycoplasma hominis (see Table 2 for detailed summary of clinical characteristics of these women). All the women had single positive cultures, except for one who had two positive blood cultures with Ureaplasma urealyticum isolated on the same collection day.

TABLE 2.

Clinical characteristics of women with genital mycoplasma bacteraemia 2008–2018.

Patient	Age (years)	Mycoplasma species present in BC	Comorbidities	Gestation	Complications	Delivery	Presenting symptoms	Other mycoplasma positive cultures	Treated with antibiotics with activity against genital mycoplasmas (if yes, Rx and duration)	Survival within 30 days (Y/N)
Bacteraemia in post‐partum period
1	32	Ureaplasma urealyticum	—	Term	—	LUSCS	Lower abdominal pain	HVS	No	Y
2	23	Ureaplasma urealyticum	T1DM, smoker, Vit D deficiency, previous TOP	38 + 6/40	Secondary perineal tear with PPH	NVD	Rigours, abdominal pain	N	PO erythromycin 500 mg QID for 7 days	Y
3	35	Ureaplasma urealyticum	Asthma	30/40	Fetal distress, chorioamnionitis	Em LUSCS	Crampy abdominal pain with malodorous yellow discharge	HVS	No	Y
4	29	Mycoplasma species and Ureaplasma urealyticum	Mild tricuspid regurgitation	36/40	Superficial abdominal collection requiring drainage *, neonatal death secondary to E coli sepsis (Day 2 PP)	Em LUSCS	Fever, rigours, lower abdominal pain	HVS	PO erythromycin 250 mg QID for 14 days	Y
5	37	Ureaplasma urealyticum	Previous miscarriage	39/40	EmLUSCS for FTP (IVF pregnancy), GA 39/40 LUSCS wound infection	Em LUSCS	Subjective fever, abdominal pain, increased PV bleeding	HVS	No	Y
6	28	Ureaplasma urealyticum	—	40/40	—	NVD	Fever	N	PO erythromycin 250 mg TDS for 7 days	Y
7	31	Ureaplasma urealyticum	Previous TOP	37/40	Placenta praevia, PPH	LUSCS	Fever, lethargy, light PV bleeding	N	PO azithromycin 500 mg stat	Y
8	29	Mycoplasma hominis	—	33/40	DCDA twins	Em LUSCS	Fever, abdominal pain, malodorous PV discharge	HVS	No	Y
9	31	Mycoplasma hominis and Ureaplasma parvum	IVDU, smoker, HCV positive, depression	24 + 6/40	FDIU	NVD	TPL with heavy PV loss due to APH	HVS	No	Y
10	32	Mycoplasma hominis and Ureaplasma parvum	Previous TOP, Vit D deficiency, GDM	31/40	MROP Perineal tear FDIU, stillbirth twin pregnancy	NVD	IOL following MCDA twin pregnancy complications	HVS, placenta	PO doxycycline 100 mg BD for 14 days	Y
11	20	Mycoplasma hominis	Previous TOP	38/40	—	NVD	Fever, rigours, abdominal pain	HVS	IV clindamycin 600 mg TDS for 3 days, then PO clindamycin 300 mg QID for total 6 weeks	Y
12	40	Mycoplasma hominis	Previous MC, GDM, Vit D deficiency	38/40	—	NVD	Fever, rigours, dysuria, urinary frequency	N	No	Y
13	32	Ureaplasma urealyticum	Obesity, PIH, GDM, Vit D deficiency, hypothyroidism	38 + 5/40	Failed instrumental delivery, Meconium liquor, PPH	Em LUSCS	Hypertension (Day 7 postpartum)	Not tested	IV azithromycin 500 mg BD for 5 days	Y
14	28	Ureaplasma parvum	—	40/40	Obstructed labour, chorioamnionitis, PPH	Em LUSCS	Fever, chills, unwell, lower abdominal pain	HVS	PO erythromycin 250 mg QID for 10 days	Y
15	23	Mycoplasma hominis and Ureaplasma spp.	Previous TOP, GBS carrier	40/40	FTP, PPH, endometritis	Em LUSCS	Fever, chills, rigours	Not tested	PO azithromycin 500 mg daily for 5 days	Y
16	33	Mycoplasma hominis	Previous MTOP and ectopic pregnancy requiring R sided salpingectomy	38/40	L salpingectomy	Em LUSCS	Fever/chills, unwell, sore throat, runny nose, mild headache, mild PV loss	*		Y
17	41 (2011)	Ureaplasma urealyticum	Vitamin D deficiency	38/40	Failed VBAC	Em LUSCS	Fever	*	No	Y
18	34	Ureaplasma parvum	—	37 + 1/40	Mild PET	Em LUSCS	Fever, rigours	*	PO roxithromycin 150 mg 7 days	Y
19	32	Ureaplasma spp.	—	38/40	? endometritis	Em LUSCS	Fever, tachycardia	*	PO azithromycin 500 mg daily for 7 days	Y
20	31	Ureaplasma spp.	—	40/40	Obstructed labour	Em LUSCS	Fetal tachycardia	*		Y
Bacteraemia associated with early pregnancy loss
21	30	Ureaplasma parvum and Ureaplasma urealyticum	Glomerulonephritis with nephrotic syndrome	8 + 4/40	Missed miscarriage	Suction curettage	Fever	*	PO erythromycin 500 mg BD for 7 days	Y
22	23	Mycoplasma hominis	—	6/40	Missed miscarriage	Suction curettage	Fevers, chills, abdominal cramps, PV bleeding	HVS	N	Y
23	17	Ureaplasma urealyticum	—	9/40	Miscarriage with septic abortion of complete molar pregnancy	—	Pelvic pain, PV clots	No	PO erythromycin 400 mg TDS for 10 days	Y
24	20	Ureaplasma spp.	Recent surgical TOP	7/40	RPOC	Suction curettage	Fever, rigours, offensive vaginal discharge	*	PO Azithromycin 1 g stat 2 doses 1 week apart	Y
25	34	Mycoplasma hominis	Recent surgical TOP	10/40	Tubo‐ovarian abscess	Suction curettage	Fever, chills, lower abdominal pain, increased PV bleeding with clots	HVS	PO doxycycline 100 mg BD 30 days	Y
26	19	Ureaplasma parvum	Recent surgical TOP	6 + 7/40	RPOC	Suction curettage	Fever, PV spotting	No	No	Y
27	30	Mycoplasma hominis	—	7 + 5/40	Ectopic pregnancy in caesarean scar	ToP	Fever, rigours	Not tested	No	Y
Gynaecological disease
28	55	Ureaplasma spp.	Metastatic ovarian cancer	n/a	Urinary retention, concurrent E. coli bacteraemia	n/a	Fever, urinary retention	*	Ciprofloxacin 500 mg bd 7 days	Y
29	30	Mycoplasma hominis	Ovarian cyst	n/a	Tubo‐ovarian abscess	n/a	Fever and lower abdominal pain	No	PO doxycycline 100 mg daily for 8 days	Y
30	52	Ureaplasma spp.	*	n/a	Coagulase negative staphylococcus concurrently isolated from blood culture	*	*	*	No	Unknown

Abbreviations: BD, bis in die (Latin), twice daily; D&C, dilation and curettage; Em, emergency; FDIU, fetal death in utero; FTP, failure to progress; GA, gestational age; GBS, group B streptococcus; GDM, gestational diabetes mellitus; HCV, hepatitis C virus; HVS, high vaginal swab; IOL, induction of labour; IV, intravenous; IVDU, intravenous drug use; IVF, in vitro fertilisation; LUSCS, lower uterine segment Caesarean section; LVS, low vaginal swab; MCDA, monochorionic diamniotic; Mh, Mycoplasma hominis ; MROP, manual removal of placenta; Ms, Mycoplasma species; MTOP, medical termination of pregnancy; N, no; n/a, not available; NVD, normal vaginal delivery; PET, pre‐eclamptic toxaemia; PO, per oral; PPH, post‐partum haemorrhage; PROM, prolonged rupture of membranes; PV, per vaginal; QID, quarter in die (Latin), four times daily; RPOC, retained products of conception; T1DM, type 1 diabetes; TDS, ter die sumendus (Latin), three times daily; TOP, termination of pregnancy; TPL, threatened pre‐term labour; VBAC, vaginal birth after caesarean section; Y, yes.

^*Insufficient documentation.

Four patients had another organism isolated from concurrently collected standard blood culture bottles. Organisms included Escherichia coli , Veillonella species, Anaerococcus prevotii , and a coagulase‐negative staphylococcus (a possible blood culture contaminant).

Of the 30 women with positive blood cultures, 20 (66%) had positive cultures collected in the post‐partum period. Seven women had cultures positive after recent termination of pregnancy or miscarriage. Two women presented with fever related to gynaecological issues (ovarian cancer, tubo‐ovarian abscess). For one woman, there were insufficient records available (study ID number 30) to determine the reason for testing.

The most common presenting symptoms were fever (n = 21), abdominal pain (n = 13), chills or rigours (n = 12), and vaginal discharge, bleeding or spotting (n = 11).

In women who had concurrent mycoplasma cultures performed from the placenta or vagina (n = 19) the same organism was isolated from both blood culture and swab in 7/19 (37%). A different genital mycoplasma was isolated from swab and blood culture in 4/19 (21%).

3.2.1. Post Partum Bacteraemia (n = 20)

12/20 (60%) women had recently undergone an emergency Caesarean section, and a further two women had elective Caesarean sections. Six women had a vaginal delivery.

In women with post‐partum fever, there was no other attributable cause found in six women, concurrent endometritis and lower respiratory tract infection in two women, and four women were diagnosed with chorioamnionitis or endometritis.

Positive blood cultures were generally taken days 0–14 post‐partum (with a median of 2 days post‐partum).

3.2.2. Bacteraemia Related to Early Pregnancy Loss (n = 7)

Seven women had bacteraemia following a recent termination of pregnancy (n = 3), ectopic pregnancy (n = 1) or miscarriage (n = 3). Six of these women presented with fever, and most had associated abdominal pain. Six women required operative intervention for their underlying condition.

3.3. Treatment and Outcomes

17/30 (57%) of women with a blood culture positive for a genital mycoplasma received antibiotics which would be expected to have some activity against the relevant mycoplasma. Several women also received a concurrent beta‐lactam antibiotic or cephalosporin for treatment of suspected chorioamnionitis/endometritis.

Regardless of the presence or absence of directed treatment, there were no patient deaths or relapse of infection in the 30‐day period following a positive blood culture for genital mycoplasma.

4. Discussion

This retrospective review of mycoplasma blood cultures collected from women attending a tertiary maternity referral hospital in Australia found that the proportion of blood culture bottles positive for a genital mycoplasma species was 0.6% over a 19‐year period, 2000–2018, with only minor fluctuations over time. This rate is lower than that reported in published data from our own institution in an earlier period (1983–1994), which reported annual blood culture positivity rates of between 0% and 10% for Mycoplasma hominis and 2.2%–16.3% for Ureaplasma urealyticum [8]. The reduction in proportion positivity may relate to changes in the patient population selected for testing over time. As previously noted, throughout the study period, mycoplasma blood culture bottles were bundled with standard aerobic and anaerobic blood culture bottles for use on hospital wards. Clinicians may have inadvertently inoculated mycoplasma blood culture bottles without having a high suspicion for mycoplasma bacteraemia, whereas the tested population may have been more selected in the 1980s.

As there are few institutions around the world that routinely make mycoplasma blood culture media available for testing maternity patients with suspected peripartum sepsis, it is difficult to find published comparator data for the blood culture positivity rate. Amongst the few published studies that have specifically looked for genital mycoplasmas on blood culture, these organisms have been detected frequently. A 1982 study of post‐partum women who had mycoplasma blood cultures collected when febrile reported a 12.8% positivity rate with Mycoplasma hominis [13]. A 1975 study by McCormack et al. reported that 26/327 (8%) of post‐partum women were bacteraemic with a genital mycoplasma [14]. These blood cultures were collected within an hour of infant delivery.

In our study, there was moderate concordance between mycoplasma blood cultures and genital tract cultures, similar to the study by McCormack et al., which reported 90% concordance between Mycoplasma hominis on blood culture and vaginal swab culture [14]. A more recent Austrian study in women undergoing elective hysterectomy reported that of 11 women with confirmed genital tract colonisation with Ureaplasma urealyticum , 4 (31%) experienced bacteraemia with Ureaplasma urealyticum in the post‐operative period [6].

Our study found that Ureaplasma spp. (68%) were more commonly isolated from blood culture than M. hominis (32%), which is similar to the ratio reported by McCormack et al. (M. hominis in 10/26 (38%) of mycoplasma blood cultures) [14].

The genital mycoplasmas frequently form part of the normal flora of the genital tract, and it remains uncertain whether detection on blood culture represents transient bacteraemia (short‐term presence of bacteria in the bloodstream that is quickly cleared by the immune system without causing persistent infection) in the peripartum setting, or whether these organisms are invasive pathogens (overcoming normal host immune defences to cause infection) that require treatment.

Transient bacteraemia is a possibility, with genitourinary procedures being classified as being of intermediate risk [17]. This has been suggested by other authors [6], and one woman in our study had two blood cultures collected 7 h apart that were positive for Ureaplasma urealyticum, but had a negative blood culture 7 h later without directed treatment. A prospective study that collects routine mycoplasma cultures from well post‐partum women would be required to properly answer this question.

Interestingly, only just over half of the women with bacteraemia in our audit were prescribed antibiotics that would be expected to have any activity against the relevant genital mycoplasma. The genital mycoplasmas, lacking a cell wall, are not susceptible to the commonly prescribed beta‐lactam antibiotics. Although formal susceptibility testing is rarely done, the genital mycoplasmas are generally susceptible to tetracyclines; additionally, Mycoplasma hominis is susceptible to clindamycin and Ureaplasma spp. are generally susceptible to macrolide antibiotics. The majority of women in the audit were prescribed oral, rather than intravenous antibiotics. This is different from prescribing patterns for standard pathogens causing bacterial sepsis, where an intravenous course of antibiotics would usually be prescribed. It is possible that the clinicians viewed bacteraemia with the genital mycoplasmas as less concerning than bacteraemia due to other organisms. or were prescribing the antibiotics as part of empiric treatment for suspected chorioamnionitis or endometritis.

Regardless of treatment status, all women survived 30 days after their positive blood culture. There were no known recurrences of infection and no hospital re‐admissions. Similarly, Naessens et al. [18] reported that untreated bacteraemia with Ureaplasma urealyticum in four afebrile post‐partum women had no adverse outcomes. Based on our audit findings, our laboratory ceased recommending routine mycoplasma blood cultures for the standard work up of suspected sepsis in the peri‐partum period.

This study has several limitations; primarily that it was retrospective in nature. Selection of patients for testing was at the discretion of the clinician, and availability of mycoplasma blood culture bottles may have influenced blood culture positivity rates over time. We could not determine whether directed treatment of mycoplasma bacteraemia impacted clinical outcome. We were also unable to compare outcomes for patients with similar presentations but negative mycoplasma blood cultures, which limits any causal interpretation.

In addition, our mycoplasma blood culture broth did not support the growth of other mycoplasma species such as M. fermentans (a coloniser of the female genital tract [19]) nor M. penetrans (which has been identified as rarely causing bacteremia in immunosuppressed persons [20]).

5. Conclusion

This dataset represents one of the largest clinical reviews of mycoplasma bacteraemia done globally. In this audit, we found that approximately 0.6% of mycoplasma blood culture bottles were positive for a genital mycoplasma species in a selected population of women at a maternity hospital.

Our clinical outcome audit found that although only just over half of women with mycoplasma detected on blood culture received antibiotic therapy which covered genital mycoplasmas, all women made a full recovery. Further prospective controlled research is required to determine whether genital mycoplasma species in blood culture require directed antibiotic treatment.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.