Accuracy of Endoscopic Ultrasound to Diagnose Nodal Invasion by Rectal Cancers: A Meta-Analysis and...

REVIEW ARTICLE – GASTROINTESTINAL ONCOLOGY

Accuracy of Endoscopic Ultrasound to Diagnose Nodal Invasionby Rectal Cancers: A Meta-Analysis and Systematic Review

Srinivas R. Puli, MD1, Jyotsna B.K. Reddy1, Matthew L. Bechtold1, Abhishek Choudhary1,

Mainor R. Antillon1, and William R. Brugge2

1Division of Gastroenterology and Hepatology, University of Missouri-Columbia, Columbia, MO; 2GI Unit, Massachusetts

General Hospital, Boston, MA

ABSTRACT

Background. Nodal staging in patients with rectal cancer

predicts prognosis and directs therapy. Published data on

the accuracy of endoscopic ultrasound (EUS) for diag-

nosing nodal invasion in patients with rectal cancer has

been inconsistent.

Aim. To evaluate the accuracy of EUS in diagnosing nodal

metastasis of rectal cancers.

Method. Study Selection Criteria: Only EUS studies

confirmed by surgical histology were selected. Data Col-

lection and Extraction: Articles were searched in Medline,

Pubmed, and CENTRAL. Statistical Method: Pooling was

conducted by both fixed-effects model and random-effects

model.

Results. The initial search identified 3610 reference arti-

cles in which 352 relevant articles were selected and

reviewed. Data were extracted from 35 studies (N = 2732)

that met the inclusion criteria. Pooled sensitivity of EUS in

diagnosing nodal involvement by rectal cancers was 73.2%

(95% confidence interval [95% CI] 70.6–75.6). EUS had a

pooled specificity of 75.8% (95% CI 73.5–78.0). The

positive likelihood ratio of EUS was 2.84 (95% CI 2.16–

3.72), and negative likelihood ratio was 0.42 (95% CI

0.33–0.52). All the pooled estimates, calculated by fixed-

and random-effect models, were similar. SROC curves

showed an area under the curve of 0.79. The P for chi-

squared heterogeneity for all the pooled accuracy estimates

was [.10.

Conclusions. EUS is an important and accurate diagnostic

tool for evaluating nodal metastasis of rectal cancers. This

meta-analysis shows that the sensitivity and specificity of

EUS is moderate. Further refinement in EUS technologies

and diagnostic criteria are needed to improve the diagnostic

accuracy.

Rectal cancer is a serious disease affecting many

patients worldwide, especially in Western Europe and

North America. In the United States in 2004, invasive

rectal cancer was found in 13.1 patients per 100,000.1,2

Rectal cancer is diagnosed in *41,000 Americans annu-

ally.1,2 Although *45% of rectal cancers are localized,

*15% have distant metastasis and *6% are not staged for

various reasons.2 Risk factors for rectal cancer include

familial polyposis syndromes (FAP, HNPCC), history of

adenomatous polyps, diabetes mellitus, obesity, excessive

alcohol, and cigarette smoking.3–7 With an increased par-

ticipation in risk factors (alcohol, smoking) and obesity in

combination with overall incidence, timely diagnosis and

adequate treatment are crucial. However, before treatment,

it is imperative to evaluate the extent of the disease with

TNM staging.

In rectal cancer, the TNM staging is usually performed

to guide treatment decisions and prognosis.8 The TNM

staging for rectal cancer is dependent on the depth of

invasion of the lesion (T), regional lymph node invasion

(N), and the presence of distant metastasis (M). N staging

depends on regional lymph node involvement. NX and N0

represent nodes that cannot be assessed or no regional

lymph node metastases, respectively. Nodal invasion by

rectal cancer is considered locally advanced disease.9

Stage 0 disease represents Tis without lymph node

involvement (N0) or distant metastasis (M0), which has the

best prognosis. Stage I disease represents T1N0M0 and

T2M0N0 lesions and correlates to a 5-year survival rate of

*85–90%.2,10–12 Stage II disease represents T3N0M0 and

� Society of Surgical Oncology 2009

First Received: 25 October 2008;

Published Online: 14 February 2009

S. R. Puli, MD

e-mail: [email protected]

Ann Surg Oncol (2009) 16:1255–1265

DOI 10.1245/s10434-009-0337-4

T4N0M0 lesions and exhibit a 5-year survival rate of

*60–65%.2,10–12 Stage III disease represents any T level

with one or two lymph nodes invaded but no distant

metastasis (T1-4N1-2M0). Stage III disease correlates with

a 5-year survival rate of *30–40%.10–12 There is a marked

decrease in survival with nodal invasion by rectal cancer.

The optimal treatment of rectal cancer varies with the

stage of disease. This difference in treatment regimens

emphasizes the importance of accurate staging. All patients

with locally advanced rectal cancer without metastasis are

treated with multimodality treatment. Surgery alone, in the

form of total pelvic exenteration, is associated with high

morbidity.13,14 Treatment with radiation followed by sur-

gery is associated with complete response in two-thirds, but

recurrence occurs in one-third of these patients.15–18

Improved rates of resectability, enhanced local control, and

increased survival can be achieved by combining moderate-

to high-dose preoperative radiation therapy with concurrent

5-FU based chemotherapy.2,20–24 Therefore, appropriate

treatment is guided by the stage of rectal cancer.

In an attempt to accurately stage rectal cancer, many

modalities have been used, from computed tomography

(CT) of the abdomen, magnetic resonance imaging (MRI)

of the abdomen, and endoscopic ultrasound (EUS).

Regarding nodal invasion, CT has a accuracy of 55–65%

and MRI has accuracy of 60–65%.25,26

With such an important difference in treatment and

prognosis between early and locally advanced disease, the

accuracy of N staging is pivotal. The reports of the accu-

racy of N staging of rectal cancers by EUS vary in the

literature. Due to this inconsistency and the clinical

importance of correct staging for treatment and prognosis,

we performed a meta-analysis to evaluate the accuracy of

EUS in staging nodal disease in rectal cancer.

METHODS

Study Selection Criteria

Only EUS studies confirmed by surgical histology were

selected. Nodal invasion of rectal cancers was defined as

invasion to perirectal lymph nodes. From this pool, only

studies from which a 2 9 2 table could be constructed for

true positive, false negative, false positive, and true nega-

tive values were included.

Data Collection and Extraction

Articles were searched in MEDLINE (through PubMed,

an electronic search engine for published articles, and

Ovid) for the years 1966 to January 2008, PubMed, Ovid

journals, EMBASE, Cumulative Index for Nursing &

Allied Health Literature, ACP journal club, DARE,

International Pharmaceutical Abstracts, old Medline,

Medline nonindexed citations, OVID Healthstar, and

Cochrane Central Register of Controlled Trials (CEN-

TRAL). The terms used for search were endoscopic

ultrasound, EUS, ultrasound, endosonography, rectal can-

cer, tumor staging, nodal invasion, staging, surgery,

sensitivity, specificity, positive predictive value, and neg-

ative predictive value. Study authors were contacted when

the required data could not be determined from the publi-

cations. The data extracted from each study were used to

construct 2 9 2 tables. Two authors (SP and JR) inde-

pendently searched and extracted the data into an

abstraction form. Any differences were resolved by mutual

agreement. The agreement between reviewers for the col-

lected data was quantified using the Cohen’s j.27

Quality of Studies

Clinical trials designed with control and treatment arms

can be assessed for quality of the study. A number of cri-

teria have been used to assess the quality of a study (e.g.,

randomization, selection bias of the arms in the study,

concealment of allocation, and blinding of outcome).28,29

There is no consensus on how to assess studies designed

without a control arm. Hence, these criteria do not apply to

studies without a control arm.29 Therefore, for this meta-

analysis and systematic review, studies were selected based

on completeness of the data and inclusion criteria. Com-

pleteness was defined as data available for true positive,

false negative, false positive, and true negative values of

the diagnostic test (EUS). Quality Assessment of Studies of

Diagnostic Accuracy Included in Systematic Reviews

(QUADAS) criteria has been proposed to evaluate quality

of diagnostic studies.30,31 This was used to evaluate the

studies on 14 items described in the QUADAS criteria.

Refining search gave352 relevant articles

35 studies met theinclusion criteria

Initial search gave3,610 potential articles

2,808 articles did notlook at rectal cancer

• 283 did not meet inclusion criteria• 14 studies did not have full data to construct 2x2 table• 20 studies were in other languages

FIG. 1 Search results

1256 S. R. Puli et al.

Statistical Methods

Meta-analysis for the accuracy of EUS in diagnosing

nodal staging of rectal cancers was performed by calcu-

lating pooled estimates of sensitivity, specificity,

likelihood ratios, and diagnostic odds ratios. EUS studies

were grouped into periods of time to standardize the

change in EUS technology, experience of endoscopist,

and EUS criteria for lymph node involvement.32 These

periods of time were 1986–1994, 1995–2000, and 2001–

2008. Pooling was conducted using both Mantel-Haenszel

Method (fixed-effects model) and DerSimonian Laird

Method (random-effects model). The confidence intervals

were calculated using the F distribution method.33 Forrest

plots were drawn to show the point estimates in each

study in relation to the summary pooled estimate. The

width of the point estimates in the Forrest plots indicates

the assigned weight to that study. For 0 value cells, a 0.5

was added as described by Cox.34 The heterogeneity of

the sensitivities and specificities were tested by applying

the likelihood ratio test.35 The heterogeneity of likelihood

ratios and diagnostic odds ratios were tested using

Cochran Q test based on inverse variance weights.36

Heterogeneity among studies was also tested using sum-

mary receiver operating characteristic (SROC) curves.

SROC curves were used to calculate the area under the

curve (AUC). The effect of publication and selection bias

on the summary estimates was tested by Egger bias

indicator and Begg-Mazumdar bias indicator.37,38 Also,

funnel plots were constructed to evaluate potential

0.60.20 0.8 1.00.4

Saitoh, et al. 1986

Pappalardo, et al. 1990

Boyce, et al. 1991

Glaser, et al. 1992

Herzong, et al. 1993

Thaler, et al. 1994

Kaneko, et al. 1995

Nielsen, et al. 1996

Sailer, et al. 1997

Osti, et al. 1997

Ramana, et al. 1997

Caseiro-Alves, et al. 1998

Massari, et al. 1998

Adams, et al. 1998

Norton, et al. 1999

Norton, et al. 1999

Spinelli, et al. 1999

Akasu, et al. 2000

Akasu, et al. 2000

Gualdi, et al 2000

Kim, et al. 2001

Garcia-Aguilar, et al. 2002

Mackay, et al. 2003

Hsieh, et al. 2003

Vanagunas, et al. 2003

Zbar, et al. 2004

Bali, et al. 2004

Rifking, et al. 2004

Manger, et al. 2004

Shami, et al. 2004

Meyenberger, et al. 2005

Maor, et al. 2006

Bianchi, et al. 2006

Kim, et al. 2006

Giovannini, et al. 2006

Pooled sensitivity = 0.73 (0.71 to 0.76)

Sensitivity

0.79

0.88

0.61

0.80

0.89

0.64

0.81

0.79

0.56

0.68

0.83

0.29

0.70

0.17

0.83

0.83

0.68

1.00

0.79

0.73

0.17

0.33

0.67

0.70

0.97

0.75

0.50

0.50

0.88

0.94

0.67

0.57

0.75

0.51

0.80

(0.67 – 0.88)

(0.47 – 1.00)

(0.39 – 0.80)

(0.68 – 0.89)

(0.79 – 0.96)

(0.35 – 0.87)

(0.63 – 0.93)

(0.63 – 0.90)

(0.21 – 0.86)

(0.46 – 0.85)

(0.36 – 1.00)

(0.04 – 0.71)

(0.54 – 0.83)

(0.02 – 0.48)

(0.52 – 0.98)

(0.52 – 0.98)

(0.48 – 0.84)

(0.69 – 1.00)

(0.69 – 0.87)

(0.39 – 0.94)

(0.00 – 0.64)

(0.24 – 0.44)

(0.57 – 0.77)

(0.51 – 0.84)

(0.84 – 1.00)

(0.53 – 0.90)

(0.21 – 0.79)

(0.30 – 0.70)

(0.83 – 0.92)

(0.70 – 1.00)

(0.09 – 0.99)

(0.29 – 0.82)

(0.35 – 0.97)

(0.34 – 0.68)

(0.63 – 0.92)

Sensitivity (95% CI)

FIG. 2 Forrest plot showing sensitivity of EUS to diagnose nodal invasion of rectal cancer

Accuracy of EUS in N Staging of Rectal Cancers: A Meta-Analysis 1257

publication bias using the standard error and diagnostic

odds ratio.39,40

RESULTS

The initial search identified 3610 reference articles.

Among these, 352 relevant articles were selected and

reviewed. Data were extracted from 35 studies (N = 2732)

that met the inclusion criteria.26,41–73 Figure 1 shows the

search results. The included 35 selected studies were

published as full-text articles in peer-reviewed journals. All

the pooled estimates given are estimates calculated by the

fixed-effect model. The change adjusted agreement analy-

sis between the reviewers for data collected separately gave

a kappa value of 1.0. QUADAS criteria to evaluate the

quality of studies showed that all the studies fulfilled 4 to 5

out of 14 described criteria.

Accuracy of EUS to N stage Rectal Cancers

Pooled sensitivity of EUS in diagnosing nodal involve-

ment by rectal cancers was 73.2% (95% CI 70.6–75.6).

Figure 2 depicts the pooled sensitivity as a Forrest plot.

EUS had a pooled specificity of 75.8% (95% CI 73.5–

78.0). Forrest plot in Fig. 3 depicts specificity from indi-

vidual studies. The positive likelihood ratio of EUS was

2.84 (95% CI 2.16–3.72), and negative likelihood ratio was

0.42 (95% CI 0.33–0.52). Figures 4 and 5 depict the pooled

positive and negative likelihood ratios as a Forrest plot.

The diagnostic odds ratio, the odds of having nodal

0.60.20 0.8 1.00.4

Saitoh, et al. 1986


Boyce, et al. 1991

Glaser, et al. 1992


Thaler, et al. 1994

Kaneko, et al. 1995


Sailer, et al. 1997

Osti, et al. 1997

Ramana, et al. 1997



Adams, et al. 1998

Norton, et al. 1999

Norton, et al. 1999


Akasu, et al. 2000

Akasu, et al. 2000

Gualdi, et al 2000

Kim, et al. 2001


Mackay, et al. 2003

Hsieh, et al. 2003


Zbar, et al. 2004

Bali, et al. 2004


Manger, et al. 2004

Shami, et al. 2004


Maor, et al. 2006


Kim, et al. 2006


Pooled specificity = 0.76 (0.74 to 0.78)

Specificity

0.14

0.83

0.91

0.83

0.27

1.00

0.89

0.56

0.91

0.64

1.00

1.00

0.84

0.97

0.68

0.53

0.79

0.60

0.72

0.80

0.33

0.82

0.90

0.77

0.94

0.67

0.65

0.92

0.65

0.32

1.00

0.80

0.81

0.61

1.00

(0.03 – 0.35)

(0.36 – 1.00)

(0.71 – 0.99)

(0.72 – 0.90)

(0.15 – 0.42)

(0.72 – 1.00)

(0.80 – 0.95)

(0.40 – 0.71)

(0.77 – 0.98)

(0.44 – 0.81)

(0.40 – 1.00)

(0.69 – 1.00)

(0.67 – 0.95)

(0.88 – 1.00)

(0.48 – 0.84)

(0.29 – 0.76)

(0.64 – 0.90)

(0.47 – 0.71)

(0.60 – 0.82)

(0.52 – 0.96)

(0.04 – 0.78)

(0.75 – 0.88)

(0.84 – 0.94)

(0.56 – 0.91)

(0.83 – 0.99)

(0.35 – 0.90)

(0.38 – 0.86)

(0.84 – 0.97)

(0.53 – 0.75)

(0.19 – 0.48)

(0.29 – 1.00)

(0.66 – 0.90)

(0.58 – 0.95)

(0.45 – 0.76)

(0.69 – 1.00)

Specificity (95% CI)

FIG. 3 Forrest plot showing specificity of EUS to diagnose nodal invasion of rectal cancer


metastasis in positive compared with negative EUS studies,

was 7.9 (95% CI 5.3–11.7). Forrest plot in Fig. 6 shows the

diagnostic odds ratios from various studies included in this

analysis. All the pooled estimates, calculated by fixed- and

random-effect models, were similar. SROC curves showed

an area under the curve of 0.79. Figure 7 shows the SROC

curve and the area under the curve. The P for chi-squared

heterogeneity for all the pooled accuracy estimates was

[0.10.

Effect of Technology over Time

EUS studies were grouped into three time periods to

standardize criteria for EUS imaging of lymph node

involvement, improvement in endoscopists experience, and

change in technology over the past 2 decades. These time

periods were 1984–1994, 1995–2000, and 2001–2008.

During these time periods, the number of studies that met

the inclusion criteria for EUS as an imaging modality only

were 6, 14, and 15, respectively. For the most recent time

period, EUS alone had a sensitivity of 70.9% (95% CI

67.3–74.3) and specificity of 78.6% (95% CI 75.5–81.5).

All pooled estimates during the three time periods are

given in Table 1. The P for chi-squared heterogeneity for

all the pooled accuracy estimates was [.10.

Bias Estimates

Publication bias calculated using Harbord-Egger bias

indicator gave a value of -0.47 (95% CI -1.79 to 0.85,

P = .52), indicating that there was no publication bias. The

Begg-Mazumdar indicator gave a Kendall tau b value of

0.01 1001

Saitoh, et al. 1986


Boyce, et al. 1991

Glaser, et al. 1992


Thaler, et al. 1994

Kaneko, et al. 1995


Sailer, et al. 1997

Osti, et al. 1997

Ramana, et al. 1997



Adams, et al. 1998

Norton, et al. 1999

Norton, et al. 1999


Akasu, et al. 2000

Akasu, et al. 2000

Gualdi, et al 2000

Kim, et al. 2001


Mackay, et al. 2003

Hsieh, et al. 2003


Zbar, et al. 2004

Bali, et al. 2004


Manger, et al. 2004

Shami, et al. 2004


Maor, et al. 2006


Kim, et al. 2006


Pooled positive LR = 2.84 (2.16 to 3.72)

Positive LR

0.91

5.25

6.70

4.55

1.22

15.20

7.17

1.79

6.48

1.90

7.86

6.88

4.47

4.83

2.59

1.76

3.24

2.36

2.81

3.64

0.25

1.86

6.66

3.02

15.84

2.25

1.42

6.33

2.48

1.38

5.00

2.80

3.94

1.32

17.42

(0.74 – 1.12)

(0.86 – 32.02)

(1.72 – 26.12)

(2.79 – 7.44)

(1.00 – 1.48)

(0.98 – 235.55)

(3.78 – 13.58)

(1.23 – 2.60)

(1.90 – 22.17)

(1.08 – 3.35)

(0.55 – 112.09)

(0.38 – 124.52)

(1.95 – 10.23)

(0.75 – 31.01)

(1.43 – 4.70)

(1.03 – 3.01)

(1.72 – 6.11)

(1.74 – 3.21)

(1.92 – 4.10)

(1.24 – 10.65)

(0.04 –1.63)

(1.19 – 2.93)

(4.15 – 10.67)

(1.45 – 6.31)

(5.28 – 47.49)

(0.98 – 5.17)

(0.60 – 3.34)

(2.68 – 14.95)

(1.84 – 3.36)

(1.08 – 1.74)

(0.34 – 74.52)

(1.37 – 5.73)

(1.50 – 10.37)

(0.80 – 2.16)

(1.16 – 262.62)

Positive LR (95% CI)

FIG. 4 Forrest plot showing positive likelihood ratio of EUS to diagnose nodal invasion of rectal cancer


0.06 (P = .62), suggesting no publication bias. The funnel

plots in Fig. 8 show no publication bias for EUS studies

estimating nodal invasion of rectal cancers.

DISCUSSION

Nodal invasion by rectal cancer is considered locore-

gional spread. Locoregional spread plays a pivotal role in

estimating survival and determining the appropriate treat-

ment. The survival is lower with locoregional spread. In

these patients, giving preoperative high-dose radiation and

chemotherapy with 5-FU-based chemotherapy improves

rates of resectability with possibly local control and

improved survival.

Apart from predicting nodal invasion, EUS also pro-

vides the ability to perform fine needle aspiration (FNA)

during the procedure. Several EUS studies report very low

(0.5–2.5%) to no complications in performing FNA.74–76

In this meta-analysis and systematic review, sensitivity

and specificity of EUS to diagnose nodal invasion by rectal

cancers was approximately 75%. On the other hand, CT

has an accuracy of 55–65% and MRI has accuracy of 60–

65%.25,26

Positive likelihood ratio of a diagnostic test is a measure

of how well the test correctly identifies a disease stage. The

higher the positive likelihood ratio, the better the diag-

nostic test performs in correctly identifying the true disease

state. On the flip side, negative likelihood ratio of a

0.01 1001

Saitoh, et al. 1986


Boyce, et al. 1991

Glaser, et al. 1992


Thaler, et al. 1994

Kaneko, et al. 1995


Sailer, et al. 1997

Osti, et al. 1997

Ramana, et al. 1997



Adams, et al. 1998

Norton, et al. 1999

Norton, et al. 1999


Akasu, et al. 2000

Akasu, et al. 2000

Gualdi, et al 2000

Kim, et al. 2001


Mackay, et al. 2003

Hsieh, et al. 2003


Zbar, et al. 2004

Bali, et al. 2004


Manger, et al. 2004

Shami, et al. 2004


Maor, et al. 2006


Kim, et al. 2006


Pooled negative LR = 0.42 (0.33 to 0.52)

Negative LR

1.56

0.15

0.43

0.25

0.40

0.38

0.22

0.38

0.49

0.50

0.24

0.72

0.36

0.86

0.25

0.32

0.41

0.08

0.30

0.34

2.50

0.81

0.36

0.39

0.03

0.38

0.77

0.54

0.19

0.20

0.43

0.54

0.31

0.80

0.22

(0.49 – 4.91)

(0.02 – 0.97)

(0.25 – 0.73)

(0.15 – 0.40)

(0.17 – 0.93)

(0.19 – 0.75)

(0.11 – 0.45)

(0.19 – 0.74)

(0.23 – 1.02)

(0.26 – 0.94)

(0.06 – 1.01)

(0.44 – 1.17)

(0.22 – 0.58)

(0.67 – 1.12)

(0.07 – 0.89)

(0.08 – 1.20)

(0.23 – 0.71)

(0.01 – 1.15)

(0.19 – 0.45)

(0.13 – 0.92)

(0.76 – 8.19)

(0.69 – 0.96)

(0.27 – 0.49)

(0.23 – 0.69)

(0.00 – 0.22)

(0.17 – 0.83)

(0.40 – 1.50)

(0.37 – 0.80)

(0.13 – 0.27)

(0.03 – 1.38)

(0.11 – 1.60)

(0.29 – 1.00)

(0.09 – 1.04)

(0.53 – 1.20)

(0.11 – 0.42)

Negative LR (95% CI)

FIG. 5 Forrest plot showing negative likelihood ratio of EUS to diagnose nodal invasion of rectal cancer


diagnostic test is a measure of how well the test correctly

excludes a disease stage. The lower the negative likelihood

ratio, the better the diagnostic test’s ability to exclude a

disease stage. EUS has a low negative likelihood ratio, but

the positive likelihood ratio is modest. So, EUS performs

better to exclude nodal invasion by rectal cancer when

compared with the ability to confirm nodal invasion. In

other words, if EUS reveals that there is no nodal invasion,

then one could be more confident in the diagnosis than if

EUS reveals there was nodal involvement.

The diagnostic odds ratio of EUS to diagnose nodal

invasion by rectal cancer is high (*8). Diagnostic odds

ratio is defined as the odds of having a positive test in

patients with true anatomic stage of the disease when

compared with patients who do not have the disease.

Clinically, if EUS demonstrates a patient with rectal cancer

has nodal metastasis, the odds of having the correct ana-

tomic stage of disease is 8 times.

The effect of FNA on the accuracy estimates could not

be evaluated as there were not enough studies with accu-

racy estimated for EUS with FNA. This is one of the

limitations of this analysis.

In all the studies included, the authors mention that the

EUS was performed by an experienced endosonographer.

EUS technologies, quality of nodal imaging, and endos-

copists’ experience have improved over the past 2 decades.

To assess the effect of these changes, studies were grouped

into periods of time. The hypothesis is that during a period

of time, the technology of EUS used might be the same. A

weakness of this kind of pooling is that some of the studies

might use older equipment, even though the paper was

published in the most recent time period. However,

0.01 1001

Saitoh, et al. 1986


Boyce, et al. 1991

Glaser, et al. 1992


Thaler, et al. 1994

Kaneko, et al. 1995


Sailer, et al. 1997

Osti, et al. 1997

Ramana, et al. 1997



Adams, et al. 1998

Norton, et al. 1999

Norton, et al. 1999


Akasu, et al. 2000

Akasu, et al. 2000

Gualdi, et al 2000

Kim, et al. 2001


Mackay, et al. 2003

Hsieh, et al. 2003


Zbar, et al. 2004

Bali, et al. 2004


Manger, et al. 2004

Shami, et al. 2004


Maor, et al. 2006


Kim, et al. 2006


Pooled diagnostic odds ratio = 7.87 (5.31 to 11.66)

Diagnostic Odds Ratio

0.59

35.00

15.56

18.49

3.06

39.73

32.87

4.74

13.33

3.83

33.00

9.55

12.46

5.60

10.56

5.56

7.98

30.97

9.47

10.67

0.10

2.30

18.34

7.67

490.67

6.00

1.83

11.67

13.40

7.00

11.67

5.20

12.75

1.65

79.80

(0.15 – 2.27)

(1.74 – 703.00)

(2.91 – 83.26)

(7.99 – 42.78)

(1.10 – 8.54)

(1.94 – 813.99)

(10.63 – 101.66)

(1.77 – 12.69)

(2.27 – 78.20)

(1.22 – 11.98)

(1.06 – 1,023.57)

(0.39 – 235.78)

(3.93 – 39.55)

(0.71 – 44.48)

(1.90 – 58.53)

(0.95 – 32.46)

(2.71 – 23.51)

(1.75 – 549.05)

(4.63 – 19.36)

(1.71 – 66.72)

(0.01 – 1.54)

(1.25 – 4.23)

(9.48 – 35.47)

(2.36 – 24.86)

(48.82 – 4,931.89)

(1.32 – 27.29)

(0.41 – 8.27)

(3.75 – 36.26)

(7.30 – 24.59)

(0.84 – 58.40)

(0.32 – 422.14)

(1.47 – 18.44)

(1.84 – 88.35)

(0.67 – 4.06)

(4.18 – 1,522.93)

Diagnostic Odds Ratio (95% CI)

FIG. 6 Forrest plot showing diagnostic odds ratio of EUS to diagnose Nodal invasion of rectal cancer


statistically, there is no alternative way of looking at this

effect, and this seems to be an accepted method of looking

at the effect of technology.32 The criteria for nodal invasion

used during the most recent time period were: larger than

1 cm, hypoechoic, or round instead of elliptical. Over the

past 2 decades, the specificity of EUS to diagnose nodal

invasion has improved, which may represent improvement

in the imaging technology, experience, or criteria. The

sensitivity of EUS to nodal invasion remains the same over

the past 2 decades.

Heterogeneity among different studies was evaluated by

the test of heterogeneity and by drawing SROC curves and

finding the AUC, since different studies might use slightly

different criteria for staging. An AUC of 1 for any diag-

nostic test indicates that the test is excellent. SROC curves

for EUS showed that the value of AUC was very close to 1,

indicating that EUS is an excellent diagnostic test for

staging nodal invasion by rectal cancers.

EUS studies with statistically significant results tend to

be published and cited. Smaller studies may show larger

treatment effects because there are fewer case-mix differ-

ences (e.g., patients with only early or late disease) than

larger trials. This publication and selection bias may affect

the summary estimates. This bias can be estimated by bias

indicators and construction of Funnel plots. Bias among

studies can affect the shape of the funnel plot. In this meta-

analysis and systematic review, bias calculations using

Egger bias indicator and Begg-Mazumdar bias indicator

showed no statistically significant bias.37,38 Furthermore,

analysis using funnel plot showed no significant publica-

tion among the studies included in this analysis.

All the studies included in this analysis were either

retrospective or consecutive. A limitation of this meta-

analysis is that there were no prospective studies in the

literature that could be included in this analysis. QUA-

DAS criteria for the quality of studies showed that the

studies fulfilled 30–35% of the 14 criteria. Some of the

criteria cannot be applied to EUS studies, so there are

some inherent weaknesses of using this kind of scoring

system to evaluate EUS studies. Though the quality of

included studies was low, all the studies in literature

evaluating nodal staging of rectal cancers had similar

scores.

This meta-analysis and systematic review was written in

accordance with the proposal for reporting by the Quality

of Reporting of Meta-analyses (QUOROM) statement.77

Since this manuscript looks at diagnostic accuracy of a test,

the study design for this meta-analysis and systematic

review conformed to the guidelines of Standards for

Reporting of Diagnostic Accuracy (STARD) initiative.78

1.0Sensitivity

0.8

0.6

0.4

0.2

0 1.00.4 0.6 0.8

Symmetric SROCAUC = 0.7985SE(AUC) = 0.0225Q* = 0.7348SE(Q*) = 0.0197

1 - Specificity0.2

FIG. 7 Forrest plot showing SROC of EUS to diagnose nodal

invasion of rectal cancer

TABLE 1 Pooled diagnostic accuracy estimates of EUS for different time periods with 95% confidence intervals

Time period No. of studies Pooled sensitivity Pooled specificity Pooled LR? Pooled LR- Pooled DOR

1984–1994 6 79.7% (74.0–84.6) 62.9% (55.5–69.9) 2.6 (1.2–5.7) 0.39 (0.25–0.610) 7.4 (1.9–28.1)

1995–2000 13 73.0% (67.9–77.7) 76.4% (72.6–79.9) 2.8 (2.2–3.6) 0.39 (0.28–0.57) 9.2 (6.5–13.1)

2001–2008 15 70.9% (67.3–74.3) 78.6% (75.5–81.5) 2.8 (1.8–4.2) 0.44 (0.30–0.64) 7.1 (3.6–14.1)

LR? positive likelihood ratio, LR- negative likelihood ratio, DOR diagnostic odds ratio

0.00

StandardError

0.75

1.50

-4 60 2 4Log(Odds Ratio)

-2

FIG. 8 Funnel plots looking at publication bias in EUS studies

looking at rectal cancer invasion of locoregional lymph nodes


CONCLUSIONS

EUS is an important and accurate diagnostic tool for

evaluating nodal metastasis of rectal cancers. This meta-

analysis shows that the sensitivity and specificity of EUS is

moderate. EUS performs better if there is no anatomic

nodal invasion than if there is anatomic nodal invasion.

Further refinement in EUS technologies and diagnostic

criteria are needed to improve the diagnostic accuracy.

REFERENCES

1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer

statistics, 2007. CA Cancer J Clin. 2007;57:43–66.

2. National Cancer Institute. Surveillance epidemiology and end results

(SEER). U.S. National Institutes of Health. http://seer.cancer.gov/

csr/1975_2004/results_merged/ sect_06_colon_rectum.pdf.

3. Atkin WS, Morson BC, Cuzick J. Long-term risk of colorectal

cancer after excision of rectosigmoid adenomas. N Engl J Med.1992;326:658–62.

4. Larsson SC, Orsini N, Wolk A. Diabetes mellitus and risk of

colorectal cancer: a meta-analysis. J Natl Cancer Inst.2005;97:1679–87.

5. Cho E, Smith-Warner SA, Ritz J, van den Brandt PA, Colditz

GA, Folsom AR, et al. Alcohol intake and colorectal cancer: a

pooled analysis of 8 cohort studies. Ann Intern Med.2004;140:603–13.

6. Giovannucci E, Ascherio A, Rimm EB, Colditz GA, Stampfer

MJ, Willett WC. Physical activity, obesity, and risk for colon

cancer and adenoma in men. Ann Intern Med. 1995;122:327–34.

7. Paskett ED, Reeves KW, Rohan TE, Allison MA, Williams CD,

Messina CR, et al. Association between cigarette smoking and

colorectal cancer in the Women’s Health Initiative. J Natl CancerInst. 2007;99:1729–35.

8. Greene FL, Page DL, Fleming ID, Fritz A, Balch CM, Haller DG,

et al. AJCC cancer staging manual. 6th ed. New York: Springer-

Verlag, 2002.

9. Guillem JG, Chessin DB, Cohen AM, Shia J, Mazumdar M,

Enker W, et al. Long-term oncologic outcome following preop-

erative combined modality therapy and total mesorectal excision

of locally advanced rectal cancer. Ann Surg. 2005;241:829–36.

10. Rich T, Gunderson LL, Lew R, Galdibini JJ, Cohen AM, Don-

aldson G. Patterns of recurrence of rectal cancer after potentially

curative surgery. Cancer. 1983;52:1317–29.

11. Minsky BD, Mies C, Recht A, Rich TA, Chaffey JT. Resectable

adenocarcinoma of the rectosigmoid and rectum: patterns of

failure and survival. Cancer. 1988;61:1408–16.

12. Willett CG, Lewandrowski K, Donelly S, Shellito PC, Convery

K, Eliseo R, et al. Are there patients with stage I rectal carcinoma

at risk for failure after abdominoperineal resection? Cancer.1992;69:1651–5.

13. Law WL, Chu KW, Choi HK. Total pelvic exenteration for

locally advanced rectal cancer. J Am Coll Surg. 2000;190:78–83.

14. Hafner GH, Herrera L, Petrelli NJ. Morbidity and mortality after

pelvic exenteration for colorectal adenocarcinoma. Ann Surg.1992;215:63–7.

15. Emami B, Pilepich M, Willett C, Munzenrider JE, Miller HH.

Effect of preoperative irradiation on resectability of colorectal

carcinomas. Int J Radiat Oncol Biol Phys. 1982;8:1295–9.

16. Dosoretz DE, Gunderson LL, Hedberg S, Hoskins B, Blitzer PH,

Shipley W, et al. Preoperative irradiation for unresectable rectal

and rectosigmoid carcinomas. Cancer. 1983;52:814–8.

17. Mendenhall WM, Million RR, Bland KI, Pfaff WW, Copeland

3rd EM. Initially unresectable rectal adenocarcinoma treated

with preoperative irradiation and surgery. Ann Surg. 1987;205:

41–4.

18. Stevens KR, Fletcher WS. High dose preoperative pelvic irradi-

ation for unresectable adenocarcinoma of the rectum or sigmoid.

Int J Radiat Oncol Biol Phys. 1983;9:148.

19. Mohiuddin M, Regine WF, John WJ, Hagihara PF, McGrath PC,

Kenady DE, et al. Preoperative chemoradiation in fixed distal

rectal cancer: dose time factors for pathological complete

response. Int J Radiat Oncol Biol Phys. 2000;46:883–8.

20. Janjan NA, Khoo VS, Abbruzzese J, Pazdur R, Dubrow R, Cleary

KR, et al. Tumor downstaging and sphincter preservation with

preoperative chemoradiation in locally advanced rectal cancer:

the M. D. Anderson Cancer Center experience. Int J Radiat OncolBiol Phys. 1999;44:1027–38.

21. Videtic GM, Fisher BJ, Perera FE, Bauman GS, Kocha WI,

Taylor M, et al. Preoperative radiation with concurrent 5-fluo-

rouracil continuous infusion for locally advanced unresectable

rectal cancer. Int J Radiat Oncol Biol Phys. 1998;42:319–24.

22. Chen ET, Mohiuddin M, Brodovsky H, Fishbein G, Marks G.

Downstaging of advanced rectal cancer following combined

preoperative chemotherapy and high dose radiation. Int J RadiatOncol Biol Phys. 1994;30:169–175.

23. Janjan NA, Crane CN, Feig BW, Cleary K, Dubrow R, Curley

SA, et al. Prospective trial of preoperative concomitant boost

radiotherapy with continuous infusion 5-fluorouracil for locally

advanced rectal cancer. Int J Radiat Oncol Biol Phys.2000;47:713–8.

24. Kwok H, Bissett IP, Hill GL. Preoperative staging of rectal

cancer. Int J Colorectal Dis. 2000;15:9–20.

25. Guinet C, Buy JN, Ghossain MA, Sezeur A, Mallet A, Bigot J-M,

et al. Comparison of magnetic resonance imaging and computed

tomography in the preoperative staging of rectal cancer. ArchSurg. 1990;125:385–8.

26. Rifkin MD, Ehrlich SM, Marks G. Staging of rectal carcinoma:

prospective comparison of endorectal US and CT. Radiology.1989;170:319–22.

27. Brennan P, Silman A. Statistical methods for assessing observer

variability in clinical measures. BMJ. 1992;304:1491–4.

28. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM,

Gavaghan DJ, et al. Assessing the quality of reports of random-

ized clinical trials: is blinding necessary? Controlled Clin Trials.1996;17:1–12.

29. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD,

Rennie D, et al. Meta-analysis of observational studies in epi-

demiology: a proposal for reporting. Meta-analysis of

observational studies in Epidemiology (MOOSE) group. JAMA.2000;283:2008–12.

30. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PMM, Kleijnen J.

The development of QUADAS: a tool for the quality assessment

of studies of diagnostic accuracy included in systematic reviews.

BMC Med Res Methodol. 2003;3:25.

31. Whiting PF, Weswood ME, Rutjes AW, Reitsma JB, Bossuyt

PMM, Kleijnen J. Evaluation of QUADAS, a tool for the quality

assessment of diagnostic accuracy studies. BMC Med ResMethodol. 2006;6:9.

32. Puli SR, Singh S, Hagedorn CH, Reddy J, Olyaee M. Diagnostic

accuracy of EUS for vascular invasion in pancreatic and pe-

riampullary cancers: a meta-analysis and systematic review.

Gastrointest Endosc. 2007;65:788–97.

33. Leemis LM, Trivedi KS. A comparison of approximate interval

estimators for the Bernoulli parameter. Am Stat. 1996;50:63–8.

34. Cox DR. The analysis of binary data. London: Methuen, 1970.

35. Agresti A. Analysis of ordinal categorical data. New York: John

Wiley & Sons, 1984.


http://seer.cancer.gov/csr/1975_2004/results_merged/

http://seer.cancer.gov/csr/1975_2004/results_merged/

36. Deeks JJ. Systematic reviews of evaluations of diagnostic and

screening tests. In Egger M, Smith GD, Altman DG (eds). Sys-

tematic reviews in health care. Meta-analysis in context. London:

BMJ Books, 2001.

37. Harbord RM, Egger M, Sterne JAC. A modified test for small-

study effects in meta-analyses of controlled trials with binary

endpoints. Stat Med. 2005;25:3443–57.

38. Begg CB, Mazumdar M. Operating characteristics of a rank

correlation test for publication bias. Biometrics. 1994;50:1088–

1101.

39. Sterne JAC, Egger M, Davey-Smith G. Investigating and dealing

with publication and other biases in meta-analysis. Br Med J.2001;323:101–5.

40. Sterne JAC, Egger M. Funnel plots for detecting bias in meta-

analysis: guidelines on choice of axis. J Clin Epidemiol.2001;54:1046–55.

41. Adams DR, Blatchford GJ, Lin KM, Ternent CA, Thorson AG,

Christensen MA. Use of preoperative ultrasound staging for

treatment of rectal cancer. Dis Colon Rectum. 1999;42: 159–66.

42. Akasu T, Kondo H, Moriya Y, Sugihara K, Gotoda T, Fujita S,

et al. Endorectal ultrasonography and treatment of early stage

rectal cancer. World J Surg. 2000;24:1061–8.

43. Akasu T, Sugihara K, Moriya Y, Fujita S. Limitations and pitfalls

of transrectal ultrasonography for staging of rectal cancer. DisColon Rectum. 1997;40:S10–5.

44. AP Zbar. Endorectal ultrasonography in rectal cancer: a preliminary

Barbadian experience. West Indian Med J. 2006;55:313–8.

45. Vanagunas A, Lin DE, Stryker SJ. Accuracy of endoscopic

ultrasound for restaging rectal cancer following neoadjuvant

chemoradiation therapy. Am J Gastroenterol 99:109–12.

46. Bali C, Nousias V, Fatouros M, Stefanou D, Kappas AM.

Assessment of local stage in rectal cancer using endorectal

ultrasonography (EUS). Tech Coloproctol. 2004;8:S170–3.

47. Bianchi P, Ceriani C, Palmisano A, Pompili G, Passoni GR,

Rottoli M, et al. A prospective comparison of endorectal ultra-

sound and pelvic magnetic resonance in the preoperative staging

of rectal cancer. Ann Ital Chir. 2006;77:41–6.

48. Boyce GA, Sivak MV Jr, Lavery IC, Fazio VW, Church JM,

Milsom J, et al. Endoscopic ultrasound in the pre-operative

staging of rectal carcinoma. Gastrointest Endosc. 1992;38:468–

71.

49. Giovannini M, Bories E, Pesenti C, Moutardier V, Moutardier V,

Lelong B, Delpero JR. Three-dimensional endorectal ultrasound

using a new freehand software program: results in 35 patients

with rectal cancer. Endoscopy. 2006;38:339–43.

50. Glaser F, Kuntz C, Schlag P, Herfarth C. Endorectal ultrasound

for control of preoperative radiotherapy of rectal cancer. AnnSurg. 1993;217:64–71.

51. Gualdi GF, Casciani E, Guadalaxara A, d’Orta C, Polettini E,

Pappalardo G. Local staging of rectal cancer with transrectal

ultrasound and endorectal magnetic resonance imaging: com-

parison with histologic findings. Dis Colon Rectum.2000;43:338–45.

52. Herzog U, von FM, Tondelli P, Schuppisser JP. How accurate is

endorectal ultrasound in the preoperative staging of rectal cancer?

Dis Colon Rectum. 1993;36:127–34.

53. Hsieh PS, Changchien CR, Chen JS, Tang R, Chiang JM, Yeh

CY, et al. Comparing results of preoperative staging of rectal

tumor using endorectal ultrasonography and histopathology.

Chang Gung Med J. 2003;26:474–8.

54. Kaneko K, Boku N, Hosokawa K, Ohtsu A, Fujii T, Koba I, et al.

Diagnostic utility of endoscopic ultrasonography for preoperative

rectal cancer staging estimation. Jpn J Clin Oncol. 1996;26:30–5.

55. Kim JC, Kim HC, Yu CS, Han KR, Kim JR, Lee KH, et al.

Efficacy of 3-dimensional endorectal ultrasonography compared

with conventional ultrasonography and computed tomography in

preoperative rectal cancer staging. Am J Surg. 2006;192:89–97.

56. Mackay SG, Pager CK, Joseph D, Stewart PJ, Solomon MJ.

Assessment of the accuracy of transrectal ultrasonography in

anorectal neoplasia. Br J Surg. 2003;90:346–50.

57. Maor Y, Nadler M, Barshack I, Zmora O, Koller M, Kundel Y,

et al. Endoscopic ultrasound staging of rectal cancer: diagnostic

value before and following chemoradiation. J GastroenterolHepatol. 2006;21:454–8.

58. Massari M, De Simone M, Cioffi U, Gabrielli F, Boccasanta P,

Bonavina L. Value and limits of endorectal ultrasonography for

preoperative staging of rectal carcinoma. Surg Laparosc Endosc.1998;8:438–44.

59. Meyenberger C, Huch Boni RA, Bertschinger P, Zala GF, Klotz

HP, Krestin GP. Endoscopic ultrasound and endorectal magnetic

resonance imaging: a prospective, comparative study for preop-

erative staging and follow-up of rectal cancer. Endoscopy.1995;27:469–79.

60. Nielsen MB, Qvitzau S, Pedersen JF, Christiansen J. Endoson-

ography for preoperative staging of rectal tumours. Acta Radiol.1996;37:799–803.

61. Pappalardo G, Reggio D, Frattaroli FM, Oddi A, Mascagni D,

Urciuoli P, et al. The value of endoluminal ultrasonography and

computed tomography in the staging of rectal cancer: a pre-

liminary study. J Surg Oncol. 1990;43:219–22.

62. Norton SA, Thomas MG. Staging of rectosigmoid neoplasia with

colonoscopic endoluminal ultrasonography. Br J Surg.1999;86:942–6.

63. Sailer M, Leppert R, Kraemer M, Fuchs KH, Thiede A. The value

of endorectal ultrasound in the assessment of adenomas, T1- and

T2-carcinomas. Int J Colorectal Dis. 1997;12:214–9; DOI:

10.1007/s003840050092.

64. Saitoh N, Okui K, Sarashina H, Suzuki M, Arai T, Nunomura M.

Evaluation of echographic diagnosis of rectal cancer using

intrarectal ultrasonic examination. Dis Colon Rectum.1986;29:234–42.

65. Shami VM, Parmar KS, Waxman I. Clinical impact of endo-

scopic ultrasound and endoscopic ultrasound-guided fine-needle

aspiration in the management of rectal carcinoma. Dis ColonRectum. 2004;47:59–65.

66. Spinelli P, Schiavo M, Meroni E, Di Felice G, Andreola S,

Gallino G, et al. Results of EUS in detecting perirectal lymph

node metastases of rectal cancer: the pathologist makes the dif-

ference. Gastrointest Endosc. 1999;49:754–8.

67. Thaler W, Watzka S, Martin F, La Guardia G, Psenner K, Bonatti

G, et al. Preoperative staging of rectal cancer by endoluminal

ultrasound vs. magnetic resonance imaging. Preliminary results

of a prospective, comparative study. Dis Colon Rectum.1994;37:1189–93.

68. Ramana KN, Murthy PV, Rao KP, Bhagawanulu, Mandapal T,

Pratap B. Transrectal ultrasonography versus computed tomog-

raphy in staging rectal carcinoma. Indian J Gastroenterol.1997;16:142–3.

69. Garcia-Aguilar J, Pollack J, Lee SH, Hernandez de Anda E,

Mellgren A, Wong WD, et al. Accuracy of endorectal ultraso-

nography in preoperative staging of rectal tumors. Dis ColonRectum. 2002;45:10–5.

70. Manger T, Stroh C. Accuracy of endorectal ultrasonography in

the preoperative staging of rectal cancer. Tech Coloproctol.2004;8:s14–5.

71. Kim JC, Yu CS, Jung HY, Kim HC, Kim SY, Park SK, et al.

Source of errors in the evaluation of early rectal cancer by end-

oluminal ultrasonography. Dis Colon Rectum. 2001;44:1302–9.

72. Osti MF, Padovan FS, Pirolli C, Sbarbati S, Tombolini V, Meli C,

et al. Comparison between transrectal ultrasonography and


http://dx.doi.org/10.1007/s003840050092

computed tomography with rectal inflation of gas in preoperative

staging of lower rectal cancer. Eur Radiol. 1997;7:26–30.

73. Caseiro-Alves F, Goncalo M, Cruz L, Ilharco J, Leite J, Agost-

inho A, et al. Water enema computed tomography (WE-CT) in

the local staging of low colorectal neoplasms: comparison with

transrectal ultrasound. Abdom Imaging. 1998;23:370–4.

74. Wiersema MJ, Vazquez-Sequeiros E, Wiersema LM. Evaluation

of mediastinal lymphadenopathy with endoscopic US-guided

fine-needle aspiration biopsy. Radiology. 2001;219:252–7.

75. Vilmann P. Endoscopic ultrasonography-guided fine-needle

aspiration biopsy of lymph nodes. Gastrointest Endoscopy.1996;43:S24–9.

76. Silvestri GA, Hoffman BJ, Bhutani MS, Hawes RH, Coppage L,

Sanders-Cliette A, et al. Endoscopic ultrasound with fine-needle

aspiration in the diagnosis and staging of lung cancer. AnnThorac Surg. 1996;61:1441–5; discussion 1445–6.

77. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF.

Improving the quality of reports of meta-analyses of randomised

controlled trials: the QUOROM statement, Lancet. 1999;354:

1896–900.

78. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP,

et al. Standards for Reporting of Diagnostic Accuracy Group.

Towards complete and accurate reporting of studies of diagnostic

accuracy: the STARD initiative. The Standards for Reporting of

Diagnostic Accuracy Group. Croat Med J. 2003;44:635–8.


Accuracy of Endoscopic Ultrasound to Diagnose Nodal Invasion by Rectal Cancers: A Meta-Analysis and...

Documents

Transcript of Accuracy of Endoscopic Ultrasound to Diagnose Nodal Invasion by Rectal Cancers: A Meta-Analysis and...