Hysteroscopy in daily practice
Dongen, H. van
Citation
Dongen, H. van. (2009, February 26). Hysteroscopy in daily practice.
Retrieved from https://hdl.handle.net/1887/13533
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13533
Note: To cite this publication please use the final published version (if applicable).
Chapter 5
Diagnostic hysteroscopy in abnormal uterine bleeding: a systematic review and meta-analysis
Heleen van Dongen Cor D. de Kroon Cathrien E. Jacobi J. Baptist M.Z. Trimbos Frank Willem Jansen Adapted from BJOG 2007;114:664-675
Chapter 5 42
Introduction
Abnormal uterine bleeding in pre- and postmenopausal women is the single most common reason for gynaecological referrals. In more than 40% of the referred patients polyps and myomas have been reported [Emanuel 1995]. The ultimate gold standard in uterine cavity evaluation is hysterectomy. However, this can not be used as a diagnostic tool. Hysteroscopy allows for direct visualisation of the cervical canal and uterine cavity, enabling observation of intrauterine abnormalities. An accurate diagnosis may result in surgical or medical treatment directed at the specific pathology and may avoid the need for major surgery. Since Gimpelson and Rappold [Gimpelson 1988] reported that hystero- scopy combined with guided biopsy was more accurate than dilatation and curettage, hys- teroscopy is considered an accurate standard in uterine cavity evaluation. Despite the lack of adequate information about the diagnostic accuracy, it is used in many studies -with and without endometrial sampling- as a reference standard [Bernard 1997; Crequat 1993;
Turner 1995]. In 2002 a high quality review was published about the accuracy of hystero- scopy, though it focused exclusively on studies reporting on the presence or absence of (pre-)malignant disorders of the endometrium [Clark 2002b]. In 2003 a systematic review and meta-analysis of the accuracy of hysteroscopy was conducted in the assessment of intracavitary abnormalities in general in premenopausal women with abnormal uterine bleeding [Farquhar 2003]. In this review, however, only studies written in English were included, and because of heterogeneity between studies no positive likelihood ratio was calculated. Therefore, the purpose of this systematic review and meta-analysis was to evaluate appropriately, without language restrictions, the diagnostic accuracy of hystero- scopy in the evaluation of intrauterine abnormalities in pre- and postmenopausal women with symptoms of abnormal uterine bleeding.
Methods
This review focused on studies in which the results of diagnostic hysteroscopy in the evaluation of the uterine cavity were compared to histology. The population of interest were pre- or postmenopausal women with symptoms of abnormal uterine bleeding. The main outcome measured in our systematic review and meta-analysis was the accuracy, by means of likelihood ratios and post-test probabilities. The secondary outcome mea- sured was the feasibility by means of technical success rates and complication rates, and the accuracy of hysteroscopy in the diagnosis of endometrial polyps and submucous myomas.
Search strategy
Electronic databases (MEDLINE, EMBASE, Current Contents, Science Citation Index and the Cochrane database) were searched for studies published between January 1st 1965 (first Medline citations) through January 1st 2006 without language selection as sug- gested for an accurate literature search [Deville 2002]. A search strategy for articles
reporting on diagnostic test evaluation was performed as proposed in the literature [Deville 2000; Haynes 1994]. The medical subject heading (MeSH) and text words for the following terms were used in the search strategy: hysteroscopy, diagnosis, histology, histopathology, hysterectomy, biopsy, sensitivity and specificity (appendix A, search strategy). Reference lists of the included studies were crosschecked in search of addi- tional relevant studies not detected by the literature search.
Selection criteria
Two authors (HvD and CdK) read the abstracts of all retrieved studies and decided on their inclusion. In case of disagreement the study was included for full reading.
The inclusion criteria were: report on the accuracy of diagnostic hysteroscopy in patients with abnormal uterine bleeding compared to histology collected with guided biopsy during hysteroscopy, operative hysteroscopy or hysterectomy as reference standard.
Studies on both pre- and postmenopausal women were eligible. The study population was considered pre- or postmenopausal if more than 70% of the patients were pre- menopausal or postmenopausal respectively. If the study population could not be clas- sified into either category, the study was excluded for subgroup analysis for this item as discussed further on in the method section. Studies in which the population contained less than 70% of patients with complaints of abnormal uterine bleeding were excluded.
Studies where more than 5% of the cases used tamoxifen or if fertility problems were the primary reason for hysteroscopy, were excluded because of different prevalence and pathophysiology influencing outcome measures [de Kroon 2003a]. Further, studies only reporting on malignancy of the endometrium were excluded. This study focused on diagnostic hysteroscopy; therefore studies reporting on diagnostic accuracy based on findings by hysteroscopy specifically designed for therapy were excluded as well.
Quality assessment
The methodological quality of each selected paper was assessed independently by the two reviewers. An adjusted version of the quality assessment tool as suggested by Deville et al. [Deville 2002] containing criteria assessing internal and external validity was used (appendix B). The internal validity criteria refer to study characteristics. External validity criteria were used to provide insight into the generalisability of the studies. The quality assessment tool was piloted in a subset of included studies and tested for reliability by means of repeatability of its use and agreement by the two reviewers. Disagreement con- cerning the quality of the included studies was solved by consensus. Since no evidence existed on the interpretation of the quality of validity tools, we decided to consider studies to be of high quality if more than or equal to ten points (>two-third of maximum score) was scored. Studies with a quality score between six and nine were considered of moderate quality, and studies with a score smaller than or equal to five were considered of low quality. The mean quality of all included studies was calculated.
Chapter 5 44
Statistics
We used guidelines for meta-analysis of diagnostic trials as published by others [Deville 2002; Irwig 1994; Khan 2001; Midgette 1993]. Outcome was defined in terms of the pres- ence of an intracavitary abnormality or a normal uterine cavity. Intracavitary abnormali- ties included all intrauterine polyps, myomas, synechiae, septae, and (pre)malignancies.
A uterine cavity was considered normal in case of functional or atrophic endometrium.
Data were abstracted as 2x2 tables of the hysteroscopy result (positive or negative for intrauterine abnormality) and the histological result (normal or abnormal) independently by the two reviewers. To define test errors, cases in which the hysteroscopy result was negative for intrauterine abnormalities and the reference standard result was abnormal were regarded as false-negative results. False-positive results were cases in which the hysteroscopy result was abnormal and the reference standard was normal.
Information on menopausal state, the number of women recruited and technical details pertaining to hysteroscopic examination was retrieved from the articles as well.
When necessary, authors were contacted and asked to supply additional information.
Furthermore, separate 2x2-tables were constructed to analyse the diagnostic accuracy specifically for benign intrauterine disorders, such as endometrial polyps and submu- cous myomas. When 2x2-tables contained empty cells, 0.5 was added to each cell to enable calculations. The sensitivity and specificity with 95% confidence intervals (95%- CI) were calculated from each 2x2-table for all included studies. Chi-square statistics, weighted for sample size, were used to evaluate heterogeneity. Subgroup analyses were performed in order to obtain the largest population with homogenous data to perform the meta-analysis. Subgroups were defined prior to the analysis according to items on the quality assessment tool, the quality of the studies, menopausal state, setting and per- formance of the procedure. The pooled sensitivity, specificity, likelihood ratios and post- test probabilities (with 95%-CI) were estimated by the fixed effects model whenever het- erogeneity could be rejected (χ2: p≥ 0.05). The post-test probabilities were derived from the likelihood ratios and the pre-test probabilities from the prevalence of intrauterine abnormalities among included studies. Likelihood ratios greater than 5 or less than 0.2 may expect to generate moderate to large conclusive shifts from pre-test to post-test probability [Jaeschke 1994]. In the presence of heterogeneity across studies, a random effects model was used. This approach produces wider 95%-CI’s, and allows for a more conservative interpretation of the results [Deville 2002].
Spearman’s correlation of sensitivities and specificities was calculated to assess whether studies originated from one-receiver-operator-characteristics-curve (ROC-curve). If Spearman’s correlation coefficient was <-0.25, studies were plotted in a sensitivity- specificity-area and a summary-ROC-curve was constructed.
Hysteroscopic procedures that failed to make a final diagnosis because of technical aspects, inadequate visualisation or patient factors were categorised as failed proce- dures. Complication rates were recorded as well.
Results
Of 409 retrieved abstracts, 55 relevant articles were selected for full reading. From crosschecking reference lists of relevant studies, another sixteen studies were included, leading to 71 articles for full reading. Fourteen authors were contacted and asked for additional data, three responded whereas eleven did not respond and had to be excluded. In total, 54 studies were excluded because of various reasons as shown in figure 1. Finally seventeen studies with 4208 procedures remained and were included in our meta-analysis. The characteristics of these studies are shown in table 1 and the sensitivity-specificity plots are displayed in figure 2a and 2b. The mean quality of the included studies was 8.7 (median 8, range 4 to 13, maximum possible score 14). The intra-observer and inter-observer agreement of the quality assessment (intra-class cor- relation alpha with 95%-CI) was 0.77 (95%-CI 0.71-0.84) and 0.72 (95%-CI 0.65-0.79) respectively.
Meta-analysis
The results of the heterogeneity analysis are shown in table 2. Five studies that included patients with postmenopausal bleeding were found to have homogeneous data [Cohen 1994; Loverro 1999; Ludwin 2003; Paschopoulos 2001; Sousa 2001]. In this group the pooled sensitivity of diagnostic hysteroscopy in the assessment of the uterine cav- ity was 0.96 (95%-CI 0.93-0.99) and the pooled specificity 0.90 (95%-CI 0.83-0.95).
The positive and negative likelihood ratios were 7.9 (95%-CI 4.79-13.10) and 0.04 (95%-CI 0.02-0.09) respectively. The pre-test probability (prevalence) of uterine cavity abnormalities in this subgroup was 0.61 (95%-CI 0.25-0.97) and changed to post-test probabilities of 0.93 (95%-CI 0.88-0.95) and 0.06 (95%-CI 0.03-0.13) for positive and negative results respectively. A summary-ROC curve could not be constructed because Spearman’s correlation coefficient of the sensitivities and specificities was -0.19.
Heterogeneity remained within the other specified groups, we therefore decided to pool all studies included by using the random effects model. The prevalence of intrauterine abnormalities was 46.6% (95%-CI 22-67%). The pooled likelihood ratios of all studies included were 6.5 (95%-CI 4.1-10.4) and 0.08 (95%-CI 0.07-0.10) for positive and negative results respectively. The pre-test probability of 0.46 changed to a post-test prob- ability of 0.78 (95%-CI 0.69-0.85) and 0.04 (95%-CI 0.03-0.05) for positive and nega- tive results respectively. The pooled likelihood ratios and post-test probabilities of the subgroup analyses, according to the quality assessment score and menopausal state, are detailed in table 3.
Endometrial polyps and submucous myomas
In a separate analysis we calculated the pooled sensitivity, specificity, likelihood ratios and post-test probabilities of all studies included for the diagnosis of endometrial polyps and submucous myomas with the random effects model. Unfortunately, a number of studies had to be excluded because the authors were unable to supply the raw data
Chapter 5 46
Table 1 | Methodological detail of all included studies.
Source Number of women Menopausal state Quality assessment Completeness of verification (%) No. (%) of failed procedures Reference test Inclusion criteria
1[Bedner 2001]160Pre- and postmenopausal4100unknownguided biopsy or D&CAUB, postmenopausal bleeding or suspicion of intrauterine pathology with TVU 2[Chechia 2001]84Premenopausal7100unknownhysterectomy, operative hysteroscopy or D&CAUB 3[Cicinelli 1994]50Premenopausal131000 (0)hysterectomyscheduled for hysterectomy 4[Cicinelli 1995]52Premenopausal131000 (0)hysterectomyscheduled for hysterectomy 5[Cohen 1994]15Postmenopausal7100unknownoperative hysteroscopy, guided biopsy or D&Cpostmenopausal bleeding or sonographically endometrial thickening (≥6mm) 6[Descargues 2001]38Pre- and postmenopausal81000 (0)guided biopsy or D&CAUB or postmenopausal bleeding 7[Dueholm 2001]108Premenopausal11 983 (2.8)hysterectomyscheduled for hysterectomy for benign disease 8[Garuti 1999b]*1050Pre- and postmenopausal7100unknownhysterectomy, operative hysteroscopy or D&CAUB or postmenopausal bleeding 9[Garuti 2001]1500Pre- and postmenopausal7100unknownhysterectomy, operative hysteroscopy, guided biopsy or D&Csuspicion of intrauterine pathology with TVU 10[Loverro 1999]106Postmenopausal101000 (0)guided biopsypostmenopausal bleeding
Source Number of women Menopausal state Quality assessment Completeness of verification (%) No. (%) of failed procedures Reference test Inclusion criteria
11[Ludwin 2003]47Postmenopausal91000 (0)hysterectomy, operative hysteroscopy or guided biopsy
postmenopausal bleeding and sonographically endometrial thickening (≥5mm) 12[Ossola 1999]55Pre- and postmenopausal91000 (0)operative hysteroscopy or D&CAUB or postmenopausal bleeding 13[Paschopoulos 1997]324unknown810012 (3.7)guided biopsyAUB, postmenopausal bleeding or cytologic findings suggesting endometrial pathology 14[Paschopoulos 2001]† 397Pre- and postmenopausal810018 (4.5)hysterectomy, operative hysteroscopy, guided biopsy or D&CAUB or postmenopausal bleeding 15[Schwarzler 1998]104Pre- and postmenopausal111000 (0)guided biopsyfailed medical therapy for hypermenorroe or metrorrhagia 16[Sousa 2001]88Postmenopausal10 8615 (17)guided biopsy or D&Cpostmenopausal bleeding 17[Towbin 1996]149Pre- and postmenopausal64413 (8.7)hysterectomy or operative hysteroscopyAUB or postmenopausal bleeding AUB=abnormal uterine bleeding, TVU=transvaginal ultrasound, D&C=dilatation and curettage. *Study could only supply raw data for polyps and myomas separately. †Study was split for pre-and postmenopausal women in heterogeneity analysis.
Chapter 5 48
Figure 1 | Profile and results of systematic review.
* One study [Paschopoulos 2001] was split for pre-and postmenopausal women in heterogeneity analysis of these subgroups.
Reference list
1.[Dueholm 2002], 2.[Gimpelson 1988], 3.[Barbero 1997], 4.[Fay 1999], 5.[Milingos 2005], 6.[Okeahialam 2001], 7.[Paschopoulos 2004], 8.[Paya 1998], 9.[Perez-Medina 1994], 10.[Rudigoz 1992], 11.[Saidi 1997], 12.[Salmaggi 1997], 13.[Simon 1993], 14.[Valle 1981], 15.[Alcazar 1996], 16.[Ayida 1997], 17.[Cameron 2001], 18.[de Jong 1990], 19.[de Wit 2003], 20.[Downes 1993], 21.[Garuti 1999a], 22.[Gezer 2004], 23.[Gianninoto 2003], 24.[Goncalves 1998], 25.[Haller 1996], 26.[La Sala 1986], 27.[Litta 1996], 28.[Mencaglia 1987], 29.[Mourits 1999], 30.[Piccolboni 1991], 31.[Stamatellos 2005], 32.[Stoikov 2005], 33.[Timmerman 1998], 34.[Townsend 1993], 35.[Turhan 2005], 36.[Vigada 1995], 37.[Alhan 2005], 38.[Cepni 2005], 39.[Epstein 2001], 40.[Saar 2004], 41.[Sevcik 1998], 42.[Ceci 2000], 43.[Ceci 2002], 44.[Ceci 2003], 45.[Garuti 2002a], 46.[Garuti 2002b], 47.[Garuti 2003], 48.[Garuti 2004], 49.[Azzena 1999], 50.[Birinyi 2004], 51.[Duffy 2003], 52.[Fabres 1998], 53.[Fedele 1991], 54.[Feng 2002], 55.[Bedner 2001], 56.[Chechia 2001], 57.[Cicinelli 1994], 58.[Cicinelli 1995], 59.[Cohen 1994], 60.[Descargues 2001], 61.[Dueholm 2001], 62.[Garuti 1999b], 63.[Garuti 2001], 64.[Loverro 1999], 65.[Ludwin 2003], 66.[Ossola 1999], 67.[Paschopoulos 1997], 68.[Paschopoulos 2001], 69.[Schwarzler 1998], 70.[Sousa 2001], 71.[Towbin 1996].
Possible relevant studies retrieved from electronic
databases (n=409)
Excluded based on title and abstract (n=355)
Relevant studies from crosschecking and hand
search (n=16) Relevant studies (n=55)
Relevant studies (n=71)
Excluded because of:
• Double publication (n=1)1
• Unable to construct 2x2-table (n=13)2-14
• Different gold standard (n=22)15-36
• No diagnostic hysteroscopy (n=5)37-41
• >5% Tamoxifen use (n=7)42-48
• Symptoms <70% AUB (n=6)49-54 Included studies (n=17)55-71
Premenopausal only (n=13) Postmenopausal only (n=5)*
Studies with postmenopausal bleeding (306 procedures) was
the largest population
Homogeneity analysis (χ2-test)
Systematic review of diagnostic hysteroscopy 49
Figure 2a | Sensitivity plot (with 95%-CI) and pooled results of diagnostic hysteroscopy in the diagnosis of intracavitary abnormalities.
Figure 2b | Specificity plot (with 95%-CI) and pooled results of diagnostic hysteroscopy in the diagnosis of intracavitary abnormalities.
hysteroscopy in the diagnosis of intracavitary abnormalities.
Sensitivity
0 0,2 0,4 0,6 0,8 1
[Bedner 2001] 0,96 (0,90 - 0,99) [Chechia 2001] 0,92 (0,81 - 0,98) [Cicinelli 1994] 0,96 (0,81 - 1,00) [Cicinelli 1995] 1,00 (0,72 - 1,00) [Cohen 1994] 0,92 (0,64 - 1,00) [Descarques 2001] 0,96 (0,81 - 1,00) [Dueholm 2001] 0,84 (0,69 - 0,94) [Garuti 2001] 0,94 (0,92 - 0,96) [Loverro 1999] 1,00 (0,91 - 1,00) [Ludwin 2003] 0,94 (0,81 - 0,99) [Ossola 1999] 0,96 (0,85 - 0,99) [Paschoupoulos 1997] 0,94 (0,89 - 0,97) [Paschoupoulos 2001] 0,92 (0,87 - 0,95) [Schwarzler 1998] 0,90 (0,79 - 0,97) [Sousa 2001] 0,98 (0,87 - 1,00) [Towbin 1996] 0,92 (0,82 - 0,98) Sensitivity (95% CI)
Pooled Sensitivity = 0,94 (0,92 to 0,95) Chi-square = 16,23; df = 15 (p = 0,3667) Inconsistency (I-square) = 7,6 %
) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )
Figure 2b Specificity plot (with 95%-CI) and pooled results of diagnostic hysteroscopy in the diagnosis of intracavitary abnormalities.
Specificity
0 0,2 0,4 0,6 0,8 1
[Bedner 2001] 0,78 (0,62 - 0,89) [Chechia 2001] 0,88 (0,71 - 0,96) [Cicinelli 1994] 1,00 (0,90 - 1,00) [Cicinelli 1995] 1,00 (0,91 - 1,00) [Cohen 1994] 1,00 (0,16 - 1,00) [Descarques 2001] 0,55 (0,23 - 0,83) [Dueholm 2001] 0,88 (0,77 - 0,95) [Garuti 2001] 0,87 (0,85 - 0,90) [Loverro 1999] 0,93 (0,83 - 0,98) [Ludwin 2003] 1,00 (0,72 - 1,00) [Ossola 1999] 0,63 (0,24 - 0,91) [Paschoupoulos 1997] 0,92 (0,87 - 0,96) [Paschoupoulos 2001] 0,95 (0,91 - 0,98) [Schwarzler 1998] 0,91 (0,79 - 0,98) [Sousa 2001] 0,71 (0,44 - 0,90) [Towbin 1996] 0,39 (0,17 - 0,64) Specificity (95% CI)
Pooled Specificity = 0,89 (0,87 to 0,90) Chi-square = 81,18; df = 15 (p = 0,0000) Inconsistency (I-square) = 81,5 %
) ) ) ) )) ) ) ) ) ) ) ) ) ) )
Chapter 5 50
needed to split results for endometrial polyps and submucous myomas. Twelve studies remained. For the diagnosis of endometrial polyps (n=12) the pooled sensitivity was 0.94 (95%-CI 0.92-0.96), while the specificity was 0.92 (95%-CI 0.91-0.94). For the diagnosis of submucous myomas (n=11), the pooled sensitivity was 0.87 (95%-CI 0.81- 0.92), and the specificity 0.95 (95%-CI 0.93-0.97). The corresponding likelihood ratios, post-test probabilities and prevalence are depicted in table 4.
Table 2 | Heterogeneity analysis and correlation of sensitivity and specificity of included studies.
Study selection Number of
studies
Number of procedures
Sensitivity p-value (χ2-test)
Specificity p-value (χ2-test)
Correlation (Spearman)
All 16 3158 0.367 0.000 -0.20
Outpatient hysteroscopy 10 2643 0.398 0.000 0.02
Gold standard operative hysteroscopy
or guided biopsy 13 2940 0.603 0.000 -0.17
Gold standard hysterectomy 3 218 0.081 0.001 0.98
Explicit definition of normal/abnormal 8 2269 0.627 0.000 -0.45
Avoidance of verification bias 3 218 0.081 0.001 0.93
Independent interpretation of tests 3 172 0.706 0.000 -0.08
Prospective design 13 2530 0.296 0.000 0.24
Retrospective design 3 628 0.376 0.000 0.91
In- and exclusion criteria mentioned 4 362 0.017 0.162 0.61
Premenopausal women only 8 825 0.496 0.000 0.15
Postmenopausal women only 5 306 0.389 0.084 -0.19
Patients scheduled for surgery only 3 218 0.081 0.001 0.93
Distension with normal saline 4 340 0.027 0.335 0.61
Distension with carbon dioxide 7 1007 0.609 0.000 -0.27
Follicular phase of cycle 2 101 1.000 0.000 ---
Time of verification Simultaneous 8 889 0.082 0.006 0.02
Sequential 8 2269 0.841 0.000 -0.31
Reference test similar 5 422 0.037 0.016 0.83
Quality score ≥10 5 373 0.111 0.000 0.37
Quality score ≥8 10 1016 0.097 0.000 0.16
Quality score ≥7 14 2927 0.301 0.000 -0.14
Table 3 | Results of the meta-analysis on the diagnostic accuracy of hysteroscopy in the evaluation of the uterine cavity stratified by different subgroups.
Subgroup No. of
procedures
Likelihood ratio (95%-CI)
Post-test probability (95%-CI) Positive Negative Positive Negative
All (n=16)55-61;63-71 3158 6.5
(4.1-10.4)
0.08 (0.07-0.10)
0.85 (0.78-0.90)
0.07 (0.06-0.08) Premenopausal women
(n=8)58;60;68;69;71 825 8.3
(2.9-23.9)
0.11 (0.08-0.15)
0.89 (0.73-0.96)
0.09 (0.07-0.13) Postmenopausal women*
(n=5)59;64;65;68;70 306 7.9
(4.8-13.1)
0.04 (0.02-0.09)
0.93 (0.88-0.95)
0.06 (0.03-0.13) Follicular phase of cycle
(n=2)58;60 101 10.8
(0.07-1706.7)
0.06 (0.02-0.21)
0.90 (0.06-0.99)
0.05 (0.01-0.15) Gold standard hysterectomy
(n=3)57;58;61 218 25.3
(2.8-230.6)
0.11 (0.04-0.29)
0.96 (0.71-0.99)
0.09 (0.03-0.20) Quality score ≥10
(n=5)57;58;61;69;70 373 9.0
(3.8-21.8)
0.11 (0.06-0.19)
0.89 (0.77-0.95)
0.09 (0.05-0.14)
*Fixed effects model was used to calculate pooled likelihood ratios.
Table 4 | Results of the meta-analysis on the diagnostic accuracy of hysteroscopy in the evaluation of endometrial polyps and submucous myomas by the random effects model.
Subgroup
Likelihood ratio (95%-CI)
Post- test probability
(95%-CI) Prevalence Positive Negative Positive Negative (%)
Endometrial polyps
All (n=12)55-60;62;63;65;66;69;71 12.9 (8.0-20.9)
0.09 (0.06-0.14)
0.85 (0.77-0.90)
0.04
(0.03-0.06) 29.8 Premenopausal women
(n=6)56-58;60;69;71
33.5 (8.2-136.0)
0.16 (0.09-0.28)
0.90 (0.69-0.97)
0.04
(0.02-0.07) 20.9 Postmenopausal women*
(n=2)59;65
12.0 (4.0-35.8)
0.04 (0.01-0.26)
0.90 (0.75-0.97)
0.03
(0.004-0.17) 43.5 Submucous myomas
All (n=11)55-61;65;66;69;71 24.7 (9.0-68.2)
0.16 (0.09-0.27)
0.88 (0.73-0.95)
0.05
(0.03-0.08) 23.4 Premenopausal women
(n=6)56-58;60;69;71
16.6 (5.4-50.6)
0.17 (0.08-0.35)
0.85 (0.64-0.94)
0.05
(0.03-0.10) 24.9
*Fixed effects model was used to assess pooled likelihood ratios.
Chapter 5 52
Feasibility
Failure rates were clearly reported in twelve (71%) of the seventeen studies. The overall success rate of diagnostic hysteroscopy was estimated at 96.6 % (SD 3.6%) when stud- ies with unclear reporting were excluded. For premenopausal women the success rate weighed for the number of procedures was 96.8% (SD 2.7%) which was significantly higher (p=0.002) than for postmenopausal women (success rate 95.6%; SD 6.4%). If cases were not weighed by number of procedures, there was no significant difference found between pre- and postmenopausal women. The success rate of outpatient (office) procedures was estimated at 96.1% (SD 3.8%). In 1399 procedures sixteen complica- tions were reported (1.0%; SD 1.6%; range 0 to 4%): thirteen vasovagal collapses, two false tracts and one perforation of the uterine wall.
Discussion
This systematic review and meta-analysis of diagnostic hysteroscopy for pre- and post- menopausal women with abnormal uterine bleeding provide information on a number of comparative studies of hysteroscopy and histology collected at hysterectomy, opera- tive hysteroscopy or guided biopsy as a reference test. It shows that diagnostic hystero- scopy is accurate in the diagnosis of intrauterine abnormalities and therefore clinically useful. Moreover, in accordance with earlier reports [Clark 2002b; Jansen 2000], our review confirms that diagnostic hysteroscopy is safe, with a low incidence of serious complications and a low failure rate.
The prevalence of intrauterine abnormalities in our review of women with abnormal uterine bleeding was 46.6%, which is consistent with previously published literature [Emanuel 1995]. The likelihood ratios were in the range that suggest that diagnostic hysteroscopy is useful both in predicting disease and excluding a non-diseased state [Jaeschke 1994]. A separate analysis concerning the accuracy of endometrial polyps and submucous myomas did not reveal any differences. Since missing endometrial polyps in postmenopausal women may result in undiagnosed malignant disorders, a subanalysis was performed, which showed similar results. Likewise, Clark et al. [Clark 2002b] have already proved in their meta-analysis that diagnostic hysteroscopy is accurate in the diagnosis of endometrial cancer.
It has been suggested that a thick endometrium obscures a complete view of the uter- ine cavity, which would especially hamper accurate detection of intrauterine abnor- malities [Parsons 1993]. For this reason we pooled studies that performed hysteroscopy solely in the follicular phase of the menstrual cycle. This failed to result in a significant increase of the post-test probability, though in order to exclude pregnancy at the time of hysteroscopy it might be practical to schedule hysteroscopy in the follicular phase of the cycle.
Further, although in this review we found a significantly higher success rate of diagnos-
tic hysteroscopy among premenopausal women than among postmenopausal women, this difference was only 1% and therefore not of any clinical importance.
Also noteworthy is that 22% of the studies included for full reading were obtained by crosschecking the reference lists of the included studies. Though this may imply a poor search strategy, it is more likely that these reports were poorly indexed, which is often the case for older reports on diagnostic accuracy [Deville 2002]. Moreover, all of the studies identified by crosschecking did eventually not meet our inclusion criteria, and were excluded.
As meta-analyses often include small numbers of studies the power of the Chi-square test is limited, so they are poor at detecting true heterogeneity among studies [Hardy 1998]. An alternative approach to quantify the effect of heterogeneity is the I2 index which describes the percentage of total variation across studies that is due to hetero- geneity rather than chance [Higgins 2003]. In this review the I2 index revealed no differ- ences compared to the Chi-square test for heterogeneity (data not shown).
The differences in results among the individual studies which were included give rea- son to criticise our review. Homogeneity of results from study to study is one of the criteria for a meta-analysis, but presence of inconsistency does not always invalidate a meta-analysis. In this situation, it is important to consider the possible reasons for het- erogeneity. We explored the sources of heterogeneity in accordance with the published guidelines, taking into account differences in methodological quality and study char- acteristics [Cook 1995; Deville 2002]. The quality of the studies included did indeed vary considerably. Nevertheless, subgroup analyses regarding quality revealed no spe- cific sources of heterogeneity. Yet, exploration of heterogeneity is often limited without access to individual patient data [Egger 2001].
Heterogeneity may also be caused by clinical differences [Thompson 1994]. Variations in patient population among studies can all result in different estimates of diagnostic accuracy. An explanation for these differences might be the fact that the diseased state is defined by the use of different diagnostic thresholds to define positive and nega- tive results [Lijmer 2002]. The primary outcome in our meta-analysis was the presence or absence of intrauterine abnormalities in stead of a certain threshold. Nevertheless, this is prone to a certain amount of subjectivity and could have caused heterogeneity.
Unfortunately, the definition of the diseased state was poorly reported and could not be solved since we had no access to individual data.
The number of studies included in this review was rather small. It was therefore of little use to examine sources of heterogeneity as thoroughly as possible since the number of available data points would have limited its significance. In order to be able to draw conclusions, we decided to base our inferences on the overall pooled results calculated by the random effects model [Yusuf 1991].
The potential bias due to variation in the histological specimens and lack of blinding in its assessment needs to be discussed as well. Hysterectomy specimens are regarded as the criterion standard for verification of intrauterine diseases, but exclusive use of
Chapter 5 54
this reference standard in a diagnostic study is not feasible [Clark 2002b]. Therefore it is not surprising that many included studies obtained histology by guided biopsy. If the phenomenon of an imperfect gold standard is ignored, there will be a tendency to underestimate the diagnostic performance of the investigated test [Walter 1999]. On the contrary, if a reference test is established while knowing the outcome of the index test, test accuracy is overestimated [Deville 2002].
Finally, nowadays the QUOROM (quality of reporting of meta-analyses ) and STARD (standards for reporting of diagnostic accuracy studies) statements are used to improve the quality of systematic reviews and meta-analyses [www.consort.org]. However, at the time this study was conducted, these statements were not yet commonly used, and therefore not applied.
In conclusion, this systematic review and meta-analysis give the strongest evidence to date that diagnostic hysteroscopy is both accurate and feasible in the diagnosis of intra- uterine abnormalities. Since diagnostic hysteroscopy is predominantly performed in the outpatient clinic, and therapy in an inpatient setting, an accurate diagnosis is important to direct treatment at the specific pathology and avoid needless surgery.
Appendix A | Search strategy in PubMed (MEDLINE) for publications on the evaluation of the diagnostic accuracy of hysteroscopy. This strategy was also used for EMBASE, Current Contents, Science Citation Index and the Cochrane database, adjusted according to specific requirements for the particular electronic database.
(((hysteroscopy AND diagnosis) OR (hysteroscop* AND diagnos*)) AND (“cytology”[Subheading] OR histology OR histopathology OR histolog* OR histopatholog* OR “Biopsy”[MeSH] OR biopsy OR biops*
OR “Hysterectomy”[MeSH] OR hysterectom* OR hysterectomy OR “Curettage”[MeSH] OR curettage) AND (“Sensitivity and Specificity”[MeSH] OR “Sensitivity and Specificity” OR (Sensitivity AND Specificity) OR
“Predictive Value of Tests”[MeSH] OR Predictive Value of Tests)) OR ((hysteroscopy OR hysteroscop*) AND (“cytology”[Subheading] OR histology OR histopathology OR histolog* OR histopatholog* OR “Biopsy”[MeSH]
OR biopsy OR biops* OR “Hysterectomy”[MeSH] OR hysterectom* OR hysterectomy OR “Curettage”[MeSH]
OR curettage) AND (“Sensitivity and Specificity”[MeSH] OR “Sensitivity and Specificity” OR (Sensitivity AND Specificity)))
Chapter 5 56
Appendix B | Quality Assessment Form
(Adjusted according to Devillé et al [Deville 2002]; respective points in brackets, minimum score 1 point, maximum score 14 points)
Item Measures
Criteria of internal validity
1 Reference standard hysterectomy (2)
hysteroscopy with biopsy (1) 2 Reference test similar within study one reference standard (1) 3 Definition of abnormal/normal definition of polyp and myoma (1)
4 Independent interpretation interpretation reference blinded for index (1) 5 Avoidance of verification bias verification independent of result (1)
6 Design prospective (consecutive series (1))
retrospective (0) Criteria of external validity
7 Setting mentioned information to identify setting (1) (outpatient, inpatient)
8 Failure failure rate mentioned (1)
9 Reason for referral inclusion criteria mentioned (1) exclusion criteria mentioned (1)
10 Information on index test information about procedure (in)directly available (1)
11 Demographic information age median/range (1) menopausal state (1) Total score: ≥10 high quality study
6-9 moderate quality study ≤5 low quality study
