Secondary Logo

Journal Logo

Original Studies

Comparison of Assays for the Diagnosis of Mycoplasma genitalium and Macrolide Resistance Mutations in Self-Collected Vaginal Swabs and Urine

Chernesky, Max PhD; Jang, Dan BSc; Martin, Irene BSc; Speicher, David J. PhD∗,‡; Clavio, Avery BSc; Lidder, Ravinder BSc; Ratnam, Sam PhD; Smieja, Marek MD, PhD; Arias, Manuel BSc; Shah, Anika BSc

Author Information
doi: 10.1097/OLQ.0000000000001226
  • Free

Mycoplasma genitalium (MG) is a sexually transmitted pathogen and a cause of persistent urethritis in men.1 It has been reported to be associated with a number of lower and upper genital tract conditions in women.2 Infection rates have generally been reported higher than Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) in most countries,3,4 and testing MG-positive samples has demonstrated a considerable frequency of macrolide resistance mutations (MRMs).5,6 In 2017, the Seeplex STD6 ACE (STD6) multiplex polymerase chain reaction (PCR) assay (Seegene Inc, Seoul, South Korea) was approved by Health Canada (HC) for the laboratory diagnosis of MG.7 The Aptima MG (AMG) assay (Hologic Inc., San Diego, CA) is a transcription-mediated amplification (TMA) assay performed on a Panther instrument8,9 and received US Food and Drug Administration clearance and HC approval for use on vaginal swabs (VS) and urine in 2018. The ResistancePlus MG (RPMG) assay (SpeeDx Pty. Ltd., Sydney, Australia) detects the MgPa gene and any of 5 particular mutations in the 23S rRNA gene associated with macrolide resistance.10–12 ResistancePlus MG received HC approval for use on VS in 2018.

The objectives of the study were as follows: (1) to compare 3 commercial assays to detect MG in self-collected VS and first-void urine (FVU) specimens in a street-youth population; (2) to compare MRM determined by the RPMG and an established laboratory-derived test, which was used at the Canadian National Microbiology Laboratory (NML) using MagNa Pure 96 (Roche Molecular Systems, Pleasanton, CA) extraction followed by PCR and Sanger sequencing of MG 23S rRNA (23SMGSS).13


Study Population

Self-collection of 3 VS and FVU was performed in a youth clinic by 300 women aged 16 to 24 years. Order of VS collection was randomized. Study candidates signed a research ethics board–approved consent form. Sexually active youth with or without symptoms, and not using antibiotics within the past 3 weeks were eligible.

Laboratory Methods

Vaginal swabs were collected with Food and Drug Administration– and HC-approved swabs: Aptima Vaginal Swab Collection kit (Hologic Inc.) for AMG, UTM Kit (Copan Italia SpA, Brescia, Italy) for STD6, and Multi-Collect Unisex Swab Collection Kit (Abbott Molecular, Des Plaines, IL) for RPMG into respective transport tubes containing the appropriate media. Swabs and FVU (20 mL) were collected in a randomized schedule. After transportation to the Infectious Research Laboratory at St Joseph's Healthcare Hamilton, each swab was processed into the respective assays within 24 hours. Urine samples were vortexed and processed according to the package inserts.

Aptima MG

The AMG test was performed on the automated Panther instrument (Hologic Inc.). The Aptima VS samples were loaded into the instrument containing the swabs, and 2 mL of FVU was pipetted into a specimen processing tube before loading. Aptima MG used 400 μL, all of which was extracted into the reaction. Uniquely positive samples were blindly retested in 2 confirmatory TMA laboratory-derived tests targeting alternate (ALT) regions of the 16S rRNA (ALT 1) and 23S rRNA (ALT 2) using a cutoff of 50,000 RLU on the Panther at Hologic.13

Seeplex STD6 ACE

For the VS in the UTM Collection Kit, 200 μL was transferred into a microcentrifuge tube for processing. First-void urine (1 mL) was centrifuged at 13,000g for 15 minutes; the supernatant fluid was discarded, and the pellet was resuspended in 200 μL of PBS. For both VS and FVU aliquots, total DNA was extracted using the NucliSENS easyMAG (bioMérieux, Saint-Laurent, Canada) and amplified on a thermocycler then visualized via gel electrophoresis as per the manufacturer's protocol.

ResistancePlus MG

Vaginal swabs and FVU in the multi-collect tubes were extracted on an m2000sp instrument (Abbott Molecular) and 5 μL of eluate mixed with 20 μL of Plex Master-mix from the RPMG kit and amplified on an ABI 7500 Fast platform. ResistancePlus MG used 3 fluorophore channels: (1) FAM for detection of the MgPa gene, (2) JOE to detect any of 5 mutations (A2058G, A2059G, A2058C, A2059C, and A2058T) in the 23S rRNA gene, and (3) TAMRA for detection of a heterologous extrinsic control.

23SMGSS for the detection of MRM was performed by NML on VS and FVU samples that were positive in the AMG and STD6 assays. Samples were shipped refrigerated and tested within 1 week. After extraction of 500 μL of sample, 5 μL of DNA eluate was used for amplification with subsequent Sanger sequencing of the 23S rRNA to identify MRM.14

Statistical Comparisons

For detection of MG agreement, tables were constructed of positives and negatives for each specimen type allowing for calculation of positive, negative, and percent overall agreement (OA) and κ statistics between each assay for each specimen type. Because three assays were compared, we established a rotating reference standard to calculate sensitivity, specificity, and predictive values with 95% confidence intervals, where the third test was measured against the agreement of the other two. For MRM detection agreement, tables were constructed for resistance and nonresistance outcomes between RPMG and 23SMGSS for each specimen type. Agreement, sensitivity, specificity, and predictive value estimates were calculated for RPMG detection of MRM using the results from the 23SMGSS procedure as the reference standard.


A total of 32 (10.6%) of enrolled patients had at least one symptom of vaginal itching, bleeding between periods, painful sex or urination, or a discharge. A history of infection with CT and or MG was reported for 18 (56.2%). Seven (21.9%) of 32 were currently positive for CT and or NG and 4 of them concurrently with MG. Table 1 shows the distribution of MG positives according to the presence or absence of symptoms, the specimen type, and the test used. Positives in each test and specimen type in the symptomatic women were fewer than in the asymptomatic group, but the percentage of the total numbers in the symptomatic group was higher. In the total group, the prevalence rates of MG by AMG were 19.7% (59/300) when testing VS and 17.2% (51/297) with FVU (Table 2). Comparing the results of RPMG to AMG, both assays agreed on 48 positive and 241 negative VS. Overall agreement was 96.3% (κ = 0.91). Testing FVU detected fewer infections in both assays, and OA was 93.3% (κ = 0.72), but AMG detected 20 more positives. Table 3 shows comparisons between AMG and STD6. Overall agreements were 86.7% (κ = 0.52) on VS and 87.9% (κ = 0.44) with FVU. Table 4 summarizes the comparisons between RPMG and STD6, where OAs were 90.3% (κ = 0.59) with VS and 92.6% (κ = 0.54) for FVU. An expanded subanalysis of Tables 2–4 according to symptoms revealed little if any differences in agreements.

TABLE 1 - M. genitalium Positive Results in Vaginal Swabs (VS) and First-Void Urine (FVU) Processed Through Aptima MG (AMG), ResistancePlus MG (RPMG), and STD6 Assays According to Symptoms
Symptomatic (n = 32) 10 (31%) 9 (28.1%) 8 (25%) 5 (15.6%) 8 (25%) 4 (12.5%)
Asymptomatic (n = 268) 49 (18.3%) 42 (15.7%) 40 (14.9%) 26 (9.7%) 25 (9.3%) 17 (6.3%)
% Positive 19.7% (59/300) 17.2% (51/297) 16% (48/300) 10.4% (31/297) 11% (33/300) 7.1% (21/297)

TABLE 2 - Agreement and κ Coefficient Between AMG and RPMG for the Identification of MG in VS and FVU
+ +
AMG + 48 11 59 AMG + 31 20 51
0 241 241 0 246 246
48 252 300 31 266 297
PA: 100% (48/48), NA: 96.6% (241/252) PA: 100% (31/31), NA: 92.5% (246/266)
OA: 96.3% (289/300) OA: 93.3% (277/297)
κ = 0.91 (very good) κ = 0.72 (good)
FVU indicates first-void urine; NA, negative agreement; OA, overall agreement; PA, positive agreement; AMG, Aptima MG; RPMG, ResistancePlus; VS, vaginal swab.

TABLE 3 - Agreement and κ Coefficient Between AMG and STD6 for the Identification of MG in VS and FVU
+ +
AMG + 26 33 59 AMG + 18 33 51
7 234 241 3 243 246
33 267 300 21 276 297
PA: 78.8% (26/33), NA: 87.6% (234/267) PA: 85.7% (18/21), NA: 88.0% (243/276)
OA: 86.7% (260/300) OA: 87.9% (261/297)
κ = 0.52 (moderate) κ 0.44 (moderate)
AMG indicates Aptima MG; FVU, first-void urine; MG, Mycoplasma genitalium; NA, negative agreement; OA, overall agreement; PA, positive agreement; STD6, Seeplex STD6 Ace; VS, vaginal swab.

TABLE 4 - Agreement and κ Coefficient Between RPMG and STD6 for the Identification of MG in VS and FVU
+ +
RPMG + 26 22 48 RPMG + 15 16 31
7 245 252 6 260 266
33 267 300 21 276 297
PA: 78.8% (26/33), NA: 91.8% (245/267) PA: 71.4% (15/21), NA: 94.2% (260/276)
OA: 90.3% (271/300) OA: 92.6% (275/297)
κ = 0.59 (moderate) κ = 0.54 (moderate)
FVU indicates first-void urine; MG, Mycoplasma genitalium; NA, negative agreement; OA, overall agreement; PA, positive agreement; RPMG, ResistancePlus; STD6, Seeplex STD6 Ace; VS, vaginal swab.

Table 5 shows that when using a rotating reference standard, the sensitivity and specificity estimates were 100% and 88.0% for AMG, 100% and 92.0% for RPMG, and 54.2% and 97.2% for STD6, respectively, on VS. Testing FVU, the sensitivity and specificity estimates were 100% and 88.3% for AMG, 83.3% and 94.3% for RPMG, and 48.4% and 98.9% for STD6, respectively.

TABLE 5 - Sensitivity, Specificity, and Predictive Values Estimates for AMG, RPMG and STD6 Assays Using a Rotating Reference Standard*
VS + FVU +
AMG + 26 33 59 Sensitivity: 100% (CI, 86.8–100)
Specificity: 88.0% (83.5–91.6)
PPV: 44.1% (36.4–52.1)
NPV: 100%
+ 15 33 48 Sensitivity: 100% (CI, (78.2–100)
Specificity: 88.3% (84.0–91.8)
PPV: 31.3% (24.8–38.5)
NPV: 100%
0 241 241 0 249 249
Total 26 274 300 15 282 297
+ +
RPMG + 26 22 48 Sensitivity: 100% (CI, 86.8–100)
Specificity: 92.0% (88.1–95.0)
PPV: 54.2% (44.2–63.8)
NPV: 100%
+ 15 16 31 Sensitivity: 83.3% (58.6–96.4)
Specificity: 94.3% (90.9–96.7)
PPV: 48.4% (35.9–61.2)
NPV: 98.9% (90.2–96.1)
0 252 252 3 263 266
Total 26 274 300 18 279 297
+ +
STD6 + 26 7 33 Sensitivity: 54.2% (CI, 39.2–68.7)
Specificity: 97.2% (94.4–98.9)
PPV: 78.8% (63.1–88.9)
NPV: 91.8% (89.1–93.8)
+ 15 3 18 Sensitivity: 48.4% (30.2–67.0)
Specificity: 98.9% (97.8–99.8)
PPV: 83.3% (60.1–94.2)
NPV: 94.3% (92.2–95.9)
22 245 267 16 266 282
Total 48 252 300 31 269 300
*Test being measured against the reference standard, where the number of true positives is determined by number of positives in both of the other 2 tests.
CI indicates percent confidence intervals.

We were able to retrieve a panel of 8 of 11 VS and 5 of 20 FVU samples, which were MG negative by RPMG and STD6 but positive by AMG, for confirmatory testing with Aptima ALT assays. Table 6 presents the outcome of ALT testing and MRM outcomes from the 23SMGSS procedure. There was a sufficient specimen volume for the ALT 1 assay, which failed to confirm results from VS 130 and VS 151. In the ALT 2 assay, 5 VS lacked sufficient volume. Six VS and 5 FVU were confirmed as positive in 1 or more of the ALT assays: all failed to amplify in the 23SMGSS procedure.

TABLE 6 - Results of Testing 13 Specimens Uniquely Positive in the Aptima MG (AMG) Assay With Alternate Aptima Assays and Outcomes for Macrolide Resistance After Testing by the 23S PCR and Sanger Sequencing (23SMGSS) Procedure
Sample AMG (16S rRNA) ALT 1 (16S rRNA) ALT 2 (23S rRNA) Confirmation Interpretation VS FVU
VS 056 1,819,063 1,360,319 1,112,004 POS NEG NEG
VS 073 1,286,357 163,809 ISV POS NEG ND
VS 079 1,683,154 1,380,883 990,140 POS NEG NEG
VS 112 1,755,624 1,529,749 ISV POS NEG ND
VS 128 1,880,292 777,325 1,118,526 POS NEG NEG
VS 130 665,938 9,248 ISV NEG NEG ND
VS 145 1,591,819 789,249 ISV POS NEG ND
VS 151 1,607,259 1,147 ISV NEG NEG ND
FVU 056 1,634,867 1,274,641 1,128,985 POS ND NEG
FVU 059 1,685,458 166,904 1,086,881 POS ND NEG
FVU 079 1,661,752 98,889 1,148,171 POS ND NEG
FVU 088 1,708,375 1,370,073 3,511 POS ND NEG
FVU 128 1,764,423 1,371,281 1,146,126 POS NEG NEG
ALT 1 indicates Aptima test for an alternate 16S rRNA sequence; ALT 2, Aptima test for a 23S rRNA sequence; FVU, first-void urine; ISV, insufficient sample volume; PCR, polymerase chain reaction; VS, vaginal swab.

Table 7 summarizes a comparison of MRM reported by RPMG and 23SMGSS performed on AMG samples (41 VS and 25 FVU). Overall agreements were 73.2% (30/41; κ = 0.41) with VS and 76.0% (19/25; κ = 0.52) for FVU. A similar agreement calculation for STD6 samples produced comparable agreement rates (data not shown). The discordant samples in Table 7 were confirmed by testing the companion RPMG samples in the 23SMGSS procedure (data not shown). Examination of the discordant samples showed that the 23SMGSS procedure reported 9 VS to be resistant, which RPMG reported as not resistant. Conversely the RPMG assay reported 8 specimens (2 VS and 6 FVU) as resistant, which 23SMGSS reported as wild-type (WT). Using the 23S results as the reference standard for MRM detection, the sensitivity and specificity estimates with 95% confidence intervals were 71.0% (51.9–85.9) and 80% (44.4–97.5) for VS and 100% (73.5–100) and 53.8% (25.1–80.8) for FVU. The distributions of mutation sequences reported from the 23S MGSS procedure were 24 A2058G, 3 A2058T, and 5 A2059G. Testing VS and FVU, the MRM frequency rates (numbers resistant divided by the numbers positive for MG) were 50.0% (24/48) and 58.1% (18/31), respectively, for RPMG (Tables 2, 7) compared with 52.5% (31/59) and 23.5% (12/51) for 23SMGSS performed on Aptima specimens (Tables 3, 7).

TABLE 7 - Macrolide Resistance Mutations by Resistance Plus MG Assay and Aptima MG Samples Processed in the 23SMGSS Procedure
VS RPMG R 22 2 24 PA: 71.0% (22/31)
NA: 80% (8/10)
OA: 73.2% (30/41)
κ = 0.41 (moderate)
Sens.: 71.0% (51.9–85.9)
Spec.: 80% (44.4–97.5)
PPV: 91.7% (75.7–97.5)
NPV: 47.1% (32.1–62.6)
NR 9 8 17
31 10 41
R 12 6 18 PA: 100% (12/12)
NA: 53.8% (7/13)
OA: 76.0% (19/25)
κ = 0.52 (Moderate)
Sens.: 100% (73.5–100)
Spec.: 53.8% (25.1–80.8)
PPV: 66.7% (52.6–78.2)
NPV: 100%
NR 0 7 7
12 13 25
FVU, first-void urine; κ, κ statistic; NA, negative agreement; NR, no resistance; NPV, negative predictive agreement; OA, overall agreement; PA, positive agreement; PPV, positive predictive agreement; R, resistance; Sens., sensitivity; Spec., specificity; VS, vaginal swab.


Infection rates were considerable in both symptomatic and asymptomatic women with substantial MRM frequencies. The prevalence of MG infection varied according to the specimen tested and the assay used. The AMG assay reported the highest MG detection rate of 19.7% (59/300) for VS compared with 17.2% (51/297) from testing FVU. M. genitalium prevalence rates using RPMG were 16.0% (48/300) for VS and 10.4% (31/297) for FVU compared with STD6 rates of 11.0% (33/300) and 7.1% (21/297), respectively. Analysis of MG prevalence according to symptoms indicated that assay agreements were not influenced by symptoms. MG prevalence rates in our study are higher than previous reports of 4% to 5% in Ontario women,3,15 but similar to the rates of 16.3% (13.4–19.8) reported for women in a multicenter clinical cohort study across the United States using an MG research use only test by Hologic4 and in the same population from an earlier Canadian study performed in the same clinic.16 The MG prevalence for self-collected VS and FVU was reported as 10.2% from the AMG Evaluation Study multicenter trial.8

We used a rotating reference standard to compare estimates of sensitivity, specificity, and predictive values and showed that AMG and RPMG performed on VS were 100% sensitive compared with 54.2% for STD6. Specificities were 88.0% for AMG, 92.0% for RPMG, and 97.2% for STD6, respectively, indicating the presence of extra positives and inflated sensitivities, which are consequences of creating a rotating reference standard when one of the tests has a much lower sensitivity. Table 5 also showed that the sensitivity of AMG was 100% for both VS and FVU, which is different from the findings of the AMG Evaluation Study trials8 where female FVU testing was less sensitive than VS. This may be a true difference between studies or due to differences in calculating sensitivity. More RPMG positives detected in VS than in FVU (sensitivity, 100% vs. 83.3%) reflects the test's claim for using VS but not FVU. Both specimens performed poorly in the STD6 assay.

We were able to perform confirmatory testing on 8 of 11 VS and 5 of 20 FVU retrieved from samples, which were only AMG positive (Table 2). Aptima-ALT assays confirmed all but 2 of these specimens (Table 6). The results are similar to those reported by Kirkconnell et al.13 Tabrizi et al.9 used an expanded reference standard made up of 2 of 3 other assays including Aptima-ALT and 2 other assays targeting the 16S rRNA gene and the MgPa gene. Le Roy et al.11 also used an expanded reference standard and measured AMG sensitivity and specificity of 100% and 99.1% compared with RPMG and an in house 23S rRNA fluorescence resonance energy transfer PCR performed on a variety of specimens from men and women with a low MG prevalence. Our measurements of 100% sensitivity with a rotating reference standard for RPMG are higher than 98.5% in a study comparing RPMG to a 23S rRNA laboratory reference PCR for the detection of MG in a variety of male and female samples.17 The STD6 multiplex assay detects CT, NG, MG, Trichomonas vaginalis, Mycoplasma hominis, and Ureaplasma urealyticum. There are limited published studies comparing STD6 with other assays for detection of MG. Its lower sensitivity and specificity compared with RPMG and AMG in our study may be due to a lower sensitivity and/or the inherent difficulties in reading gels with weak positive reactions. These factors may make the STD6 assay less desirable to some laboratories. The original report for detection of MG with the STD6 contained limited specimens (n = 4) and positives (n = 2) that were confirmed by a Seegene monoplex assay.7 A subsequent report confirmed 8.1% (51/626) of specimens as MG positive using a Seegene Anyplex II STI-7 assay.18

The MRM frequency estimates are the number of specimens reported resistant in Table 7 from the total number reported as MG positive in Tables 2–4. Using RPMG, the MRM frequencies were 50% (24/48) for VS and 58.1% (18/31) for FVU, similar to 63.1% reported by Tabrizi et al.10 testing urogenital and anal specimens. The 23SMGSS procedure performed on AMG-positive VS and FVU samples reported MRM frequencies of 52.5% (31/59) and 23.5% (12/51), respectively. The lower MRM rate for FVU was due to a large number of AMG-positive urines failing to amplify in the 23S PCR of the 23SMGSS procedure. The AMR multicenter cohort study4 reported an MRM frequency of 50.8% using a reverse transcription PCR of MG 23S rRNA and Sanger sequencing on cervical, vaginal, and urine specimens. The MRM values for STD6 specimens with the 23SMGSS procedure were 78.8% (26/33) for VS and 52.4% (11/21) for FVU (data not included in Table 7). Because each MG assay is created with a different level of detection, not all MG-positive specimens will be detected in the 23SMGSS system. Thus, although unique AMG-positives were confirmed by alternate TMA assays (Table 6), they were not amplifiable in the MRM PCR assay, which may be due to differences in level of detection. If these samples were excluded from the frequency denominator, a higher MRM frequency would be reported. Differences in MRM frequency rates can be influenced by WT genes and genes with resistance mutations occurring together in clinical specimens,19,20 which may explain the 2 VS and 6 FVUs (Table 7) reported as resistant by RPMG but WT by 23SMGSS. Discordancy between the 2 MRM tests may also be due to sensitivity levels and the fact that 23SMGSS identifies the mutation sequence numbers or the presence of WT genes, whereas the RPMG reports resistant or not resistant, some of which may be WT or resistance below the test cutoff, which may be an explanation for the 9 VS reported as resistant by 23SMGSS but not resistant by RPMG. The study by Le Roy et al.,11 testing samples from a non–sexually transmitted disease population with the RPMG assay reported a low frequency of 8.3% MRM from a population with a 5.9% MG prevalence. Similar to the present study, they found that the 23SMGSS assay detected more resistance mutations than RPMG. Our study found that most mutations were A2058G with fewer numbers of A2058T and A2059G. None of the previous studies did a detailed MRM agreement comparison between RPMG and 23SMGSS (Table 7). Because HC accepts the 23SMGSS procedure as the reference standard to compare other MRM detection assays, we calculated the sensitivity and specificity of RPMG. As shown in Table 7, the results are impacted by the different ways the assays determine and report MRM. The comparison calculations are influenced by limited numbers with wide confidence intervals. A much larger study may provide values for greater significance.

Our study has several limitations. M. genitalium prevalence rates and MRM frequencies in this young sexually active group of women, with a 10.6% sexually transmitted disease symptomatic rate and with rates of 15.1% for CT and 6.6% for NG,16 may not be translatable to the general population. Calculation of MG sensitivity and specificity using a rotating reference standard can generate inflated sensitivity values when 1 of the 3 tests detect substantially fewer positives. The STD6 assay was selected for this study because it was the only assay with HC approval at the time. Seegene's more recent multiplex STI 4 assay provides better sensitivity and specificity (unpublished data) and would have been a better comparator. Because the AMG TMA assay is very sensitive, it identifies more positives,3,9 which may not be amplified in the 23SMGSS procedure, although they are shown to be true positives with the ALT assays.13 This creates a problem for calculating MRM frequency rates depending on whether the MG numbers from the MG testing or those amplified in the 23SMG PCR are used in the denominator for the rate calculation.

This is the first application of the RPMG assay using the m2000sp (Abbott Molecular) for nucleic acid extraction followed by qPCR amplification on an ABI7500. Other RPMG validations have used MagNA Pure 96 (Roche) extracted samples on the Roche Light Cycler 480 II real-time PCR system (LC480)10 or on the ABI 7500 FAST platform.11 This alternate automated extraction using the m2000 allows the RPMG kit more flexibility for laboratories with an m2000 open system.

Our study showed excellent performance of the AMG and RPMG assays for the detection of MG in VS, but FVU was a less desirable specimen for RPMG. The STD6 assay demonstrated lower sensitivity with both specimen types. ResistancePlus MG demonstrated high specificity for both specimen types. Extra positives in the AMG assay were confirmed by ALT-Amp testing to be positive for MG but were not amplifiable in the 23SMGSS procedure. The MRM frequency agreement between RPMG and 23SMGSS was moderate because of assay detection cutoff differences and the possibility of both resistance and WT genes appearing together in some clinical specimens.


1. Jensen JS, Bjornelius E, Dohn B, et al. Use of TaqMan 5′ nuclease real-time PCR for quantitative detection of Mycoplasma genitalium DNA in males with and without urethritis who were attendees at a sexually transmitted disease clinic. J Clin Microbiol 2004; 42:683–692.
2. Lis R, Rowhani-Rahbar A, Manhart LE. Mycoplasma genitalium infection and female reproductive tract disease: A meta-analysis. Clin Infect Dis 2015; 61:418–426.
3. Chernesky MA, Jang D, Martin I, et al. Mycoplasma genitalium antibiotic resistance-mediating mutations in Canadian women with or without Chlamydia trachomatis infection. Sex Transm Dis 2017; 44:433–435.
4. Getman D, Jiang A, O'Donnell M, et al. Mycoplasma genitalium prevalence, coinfection, and macrolide antibiotic resistance frequency in a multicenter clinical study cohort in the United States. J Clin Microbiol 2016; 54:2278–2283.
5. Bradshaw CS, Jensen JS, Tabrizi SN, et al. Azithromycin failure in Mycoplasma genitalium urethritis. Emerg Infect Dis 2006; 12:1149–1152.
6. Jensen JS, Bradshaw CS, Tabrizi SN, et al. Azithromycin treatment failure in Mycoplasma genitalium-positive patients with nongonococcal urethritis is associated with induced macrolide resistance. Clin Infect Dis 2008; 47:1546–1553.
7. Lee SJ, Park DC, Lee DS, et al. Evaluation of Seeplex® STD6 ACE detection kit for the diagnosis of six bacterial sexually transmitted infections. J Infect Chemother 2012; 18:494–500.
8. Gaydos CA, Manhart LE, Taylor SN, et al. Molecular testing for Mycoplasma genitalium in the United States: Results from the AMES prospective multicenter clinical study. J Clin Microbiol 2019; 57:1–12.
9. Tabrizi SN, Costa AM, Su J, et al. Evaluation of the Hologic Panther transcription-mediated amplification assay for detection of Mycoplasma genitalium. J Clin Microbiol 2016; 54:2201–2203.
10. Tabrizi SN, Su J, Bradshaw CS, et al. Prospective evaluation of ResistancePlus MG, a new multiplex quantitative PCR assay for detection of Mycoplasma genitalium and macrolide resistance. J Clin Microbiol 2017; 55:1915–1919.
11. Le Roy C, Pereyre S, Henin N, et al. French prospective clinical evaluation of the Aptima Mycoplasma genitalium CE-IVD assay and macrolide resistance detection using three distinct assays. J Clin Microbiol 2017; 55:3194–3200.
12. Su JP, Tan LY, Garland SM, et al. Evaluation of the SpeeDx ResistancePlus MG diagnostic test for Mycoplasma genitalium on the applied biosystems 7500 Fast Quantitative PCR Platform. J Clin Microbiol 2018; 56:1–3.
13. Kirkconnell B, Weinbaum B, Santos K, et al. Design and validation of transcription-mediated-amplification nucleic acid amplification tests for Mycoplasma genitalium. J Clin Microbiol 2019; 57:1–9.
14. Jensen JS. Protocol for the detection of Mycoplasma genitalium by PCR from clinical specimens and subsequent detection of macrolide resistance-mediating mutations in region V of the 23S rRNA gene. Methods Mol Biol 2012; 903:129–139.
15. Gesink D, Racey CS, Seah C, et al. Mycoplasma genitalium in Toronto, Ont: estimates of prevalence and macrolide resistance. Can Fam Physician 2016; 62:e96–e101.
16. Chernesky M, Jang D, Martin I, et al. Mycoplasma genitalium, Chlamydia trachomatis, and Neisseria gonorrhoeae detected with Aptima assays performed on self-obtained vaginal swabs and urine collected at home and in a clinic. Sex Transm Dis 2019; 46:e87–e89.
17. Edberg A, Jurstrand M, Johansson E, et al. A comparative study of three different PCR assays for detection of Mycoplasma genitalium in urogenital specimens from men and women. J Med Microbiol 2008; 57:304–309.
18. Moon S, Choi JE, Park KI. Comparison of the Anyplex II STI-7 and Seeplex STD6 ACE detection kits for the detection of sexually transmitted infections. J Lab Med Qual Assur 2013; 35:87–92.
19. Forslund O, Hjelm M, El-Ali R, et al. Mycoplasma genitalium and macrolide resistance-associated mutations in the Skane region of southern Sweden 2015. Acta Derm Venereol 2017; 97:1235–1238.
20. Touati A, Peuchant O, Jensen JS, et al. Direct detection of macrolide resistance in Mycoplasma genitalium isolates from clinical specimens from France by use of real-time PCR and melting curve analysis. J Clin Microbiol 2014; 52:1549–1555.
Copyright © 2020 American Sexually Transmitted Diseases Association. All rights reserved.