Accuracy of PET/CT with FDG in mediastinal lymph node staging of patients with NSCLC

doi:10.4236/health.2010.23030

Paper Menu >>

Journal Menu >>

Vol.2, No.3, 204-210 (2010)

doi:10.4236/health.2010.23030

Health

Accuracy of PET/CT with FDG in mediastinal lymph node

staging of patients with NSCLC

E. Pelosi1, A. Billè2, A. Skanjeti3, V. Arena1, F. Ardissone2

1IRMET PET Center, Turin, Italy; e.pelosi@irmet.com

2Thoracic Surgery Unit, Department of Clinical and Biological Sciences, University of Turin, Italy

3Nuclear Medicine Unit, Department of Internal Medicine, University of Turin, Italy

Received 26 December 2009; revised 3 January 2010; accepted 7 January 2010.

ABSTRACT

PET/CT and contrast enhancement CT (CECT)

are the two fundamental non-invasive exams in

the preoperative staging of patients with non

small cell lung cancer (NSCLC). In the staging of

the mediastinum, recent studies show that PET

is more accurate than CECT, with an average

sensitivity and specificity of 85 vs. 61% and 90

vs. 79%, respectively. However, thanks to its

specificity value of 100%, the gold standard

remains the mediastinoscopy. The aim of this

study was to evaluate the PET/CT accuracy in

the intrathoracic lymph node staging. Three

hundred and five consecutive patients with

proven or suspected non-small cell lung cancer

who had an integrated PET/CT study were re-

trospectively evaluated. Lymph node staging

was pathologically confirmed on tissue speci-

mens obtained at surgery. A med-line research

of papers on accuracy of integrated PET/CT in

lymph node staging was also carried out. In this

population of patients, a total of 1972 lymph

node stations were evaluated. Integrated

PET/CT correctly staged 247 out of 305 patients:

188 of 214 (87.8%) N0 patients, 34 of 40 (85.0%)

N1 patients and 25 of 51 (49.0%) N2/N3 patients.

PET/CT understaged 32 patients (10.5%) and

overstaged 26 patients (8.5%). One hundred and

forty-three lymph nodes were proved positive

for malignancy. PET/CT correctly identified 89

metastatic lymph node stations. The overall

sensitivity, specificity, positive and negative

predictive value and accuracy of PET/CT were

64.8%, 87.9%, 69.4%, 85.5%, and 81.0% on a

per-patient basis and 62.2%, 97.9%, 69.5%,

97.1%, and 95.3%, on a per-nodal-station basis,

respectively. According to nodal size, PET/CT

correctly identified 67 out of 77 (87.0%) metas-

tatic lymph node stations with a short-axis di-

ameter ≥ 10 mm, and 22 out of 66 (33.3%) me-

tastatic lymph node stations with a short-axis

diameter < 10 mm (p<0.001). The incidence of

false negative lymph node metastases at

PET/CT was higher in patients with adenocar-

cinoma (42 out of 54). These data are in agree-

ment with the published literature and confirm

that integrated PET/CT is more accurate than CT

in detecting nodal metastases; however, the

PET/CT exam is not enough accurate to subs-

titute mediastinoscopy.

Keywords: Integrated PET/CT;

Intrathoracic Lymph Node Staging; Lung Cancer

1. INTRODUCTION

Lung cancer is still the leading cause of death among all

cancers. In 2006 more than 170,000 new cases were di-

agnosed in the USA [1]. The overall 5-year survival rate

of patients undergoing surgery is less than 50% [2]. The

successful management of these patients deals with an

accurate staging and advances in chemo-radiotherapy and

surgery.

An accurate preoperative staging is done with contrast

enhancement Computed Tomography (CECT) and 18F-

fluoro-2-deoxy-D-glucose integrated Positron Emission

Tomography and Computed Tomography (FDG-PET/CT).

Although CECT is the most commonly used non-invasive

modality for the evaluation of primary tumor characteris-

tics (i.e. size, location, and extent), a number of reviews

and meta-analyses have shown the limited reliability of

this modality in lymph node staging [3-6]. In recent years,

integrated PET/CT has become one of the most important

exams in the preoperative staging of these patients, due to

its ability in detecting distant and mediastinal lymph node

metastasis [7-10].

Mediastinal lymph node status has an important impact

on the course of therapy and prognosis of non small cell

lung cancer (NSCLC) cases. In patients with N2 disease,

neo-adjuvant chemotherapy followed by surgery or che-

E. Pelosi et al. / HEALTH 2 (2010) 204-210

205

mo-radiation therapy with curative intent are the accepted

treatments [11]. However, in mediastinal staging false

positive PET/CT results have been reported in patients

with coexistent inflammatory or infectious diseases,

while, due to a still suboptimal spatial resolution, PET/CT

study may be unable to identify metastatic deposits in

normal-sized lymph nodes [12-17]. Mediastinoscopy,

thanks to its 100% of specificity, remains the gold stan-

dard exam in mediastinal nodal staging; however, it has

several shortcomings including risk of morbidity and

mortality and costs [18]. Due to these reasons, the medi-

astinic staging algorithm combining integrated PET/ CT

and invasive procedures is still under debate.

The aim of this paper is to present our results about

PET/CT accuracy in the mediastinal nodal staging of

patients with NSCLC.

2. METHODS

2.1. Patient Characteristics and Inclusion

Criteria

From August 2004 to March 2009, 890 consecutive pa-

tients underwent surgery (mediastinoscopy, anterior me-

diastinotomy and/or thoracotomy) for suspected or

pathologically proven, clinically resectable NSCLC, in

our Thoracic Surgery Unit. Out of them, 305 patients

(34.3%) before undergoing surgery had an integrated

PET/CT. All PET studies were acquired in the same PET

center using a Discovery ST scanner (GE Medical sys-

tem), within 3 weeks from the surgical operation. Patients

who underwent preoperative chemo or radiotherapy or

who had a PET/CT negative primary tumor or who had a

PET/CT in other centers were excluded. Patient charac-

teristics are summarized in Table 1.

All the patients of the study population had a history

and physical examination, laboratory tests, spirometry,

chest X ray, contrast enhanced brain, chest and upper

abdomen CT and bronchoscopy to complete the pre-

operative diagnostic work up.

2.2. Integrated PET/CT

Before the exam, all patients provided informed written

consent. Patients fasted for at least 6 h before the exam;

the scanning was performed 60 min after the intravenous

administration of FDG (4.5 to 5.5 MBq/kg). After de-

termining the imaging field, a CT scan (140 kV, tube

current 60 mA/S) was performed and it was used for both

anatomical localization and for calculation of attenuation

correction. Then, the PET data were acquired in 3D mode

from the skull base to the pelvic floor in 8 to 9 bed posi-

tions. The acquisition time for PET was 3 minutes per bed

position. Coronal, sagittal, and transverse data sets were

reconstructed. Coregistered scans were displayed by

using dedicated software (Advantage 4.2; GE Healthcare)

Table 1. Characteristics of the study population (n=305).

Variable No. %

Gender

Male 239 78.4

Female 66 21.6

Age (years)

Mean ± SD 66.5 ± 8.4

Range 37 – 86

Site

Right 176 57.7

Left 129 42.3

Location

Central (inner 1/3 of lung

field) 100 32.8

Peripheral 205 67.2

Tumor diameter (cm)

Mean ± SD 3.3 ±1.8

Range 0.5 – 13

Histology

Squamous cell 77 25.2

Adenocarcinoma 183 60.0

Bronchoalveolar 6 2.0

Carcinoid # 15 4.9

Other type non-small cell

lung cancer 24 7.9

Grade of differentiation

Good 41 13.4

Moderate 139 45.6

Poor 125 41.0

Presence of vascular

invasion

Yes 80 26.2

No 225 73.8

Presence of necrosis

Absent 109 35.7

Focal 106 34.8

Extended 90 29.5

Tumor SUVmax

Mean ± SD 10.1 ± 9.5

Range 1.7 – 54

SD: standard deviation; SUVmax : maximum standardized uptake value

# Typical n=11; Atypical n=4

E. Pelosi et al. / HEALTH 2 (2010) 204-210

206

and integrated PET/CT data sets were prospectively

evaluated in consensus by two nuclear medicine physi-

cians (E.P. and V.A.) who were aware of clinical and

stand-alone contrast-enhanced CT results, but blinded to

the histologic findings. The maximum standardized up-

take value (SUVmax) of the primary tumor was measured

with a region-of-interest technique and calculated by the

software according to standard formulas. Pulmonary and

mediastinal lymph node stations, localised according to

the classification scheme of Mountain and Dresler [19],

were considered positive for metastatic spread if they

showed focally increased FDG uptake (ie, uptake higher

than the normal mediastinal activity).

2.3. Surgery and Histopathology

All 305 patients underwent surgical staging. Thirty five

patients underwent invasive mediastinal staging accord-

ing with the positive results of the integrated PET/CT for

N2/N3 disease. Cervical mediastinoscopy was used to

sample stations 2R, 4R, 2L, 4L, and 7, and anterior me-

diastinotomy was used to sample stations 5 and 6. Seven

patients were excluded from subsequent surgery due to

multi-station N2 disease (n=6) or N3 disease (n=1). The

other 28 patients underwent invasive mediastinal staging

procedure followed by thoracotomy during the same

surgical session due to non metastatic mediastinal lymph

nodes (n=8) or N2 minimal disease(n=20), defined as

single-station, intranodal metastatic deposit. The 270

remaining patients, considered N2 lymph node negative

by PET/CT, underwent thoracotomy, pulmonary resec-

tion, and complete thoracic lymphadenectomy. Overall,

pulmonary resections included pneumonectomy (n=19),

bilobectomy (n=9), lobectomy (n=252), and segmentec-

tomy (n=18). At thoracotomy, complete thoracic lym-

phadenectomy was routinely performed; it consisted of

en-bloc resection of all lymph nodes that were accessible

in the mediastinum and hilum. Intrapulmonary lymph

nodes (stations 11 and 12) were included in the resected

lung specimen. At the subcarinal level, the contralateral

mediastinal lymph nodes, lying on the opposite main

stem bronchus, were removed in 41 patients.

The pathologic review (primary tumor characteristics

and lymph node status) was performed by standard tech-

niques while immunohistochemistry was used when

appropriate. Pathologic TNM staging was performed and

disease was classified as stage IA in 89 patients (29.2%),

stage IB in 91 (29.8%), stage IIA in 10 (3.3%), stage IIB

in 43 (14.1%), stage IIIA in 49 (19.5%), stage IIIB in 14

and stage IV in 9 patients.

2.4. Data Analysis

The sensitivity, specificity, positive predictive value,

negative predictive value, and accuracy of integrated

PET/CT in the assessment of intrathoracic lymph node

involvement were determined by using histologic results

as reference standard. Diagnostic characteristics of inte-

grated PET/CT were assessed on a per-patient basis and

on a per-nodal-station basis. A probability value of <

0.05 was considered statistically significant. Statistical

analysis was carried out with StatSoft version 6.1 soft-

ware.

Finally, the results of our experience were compared

with that of the literature. With the help of the Medline

search engine PubMed, we reviewed the literature be-

tween 2000 and 2009. The search was performed using

keywords such as positron emission tomography, patho-

logical staging and lung cancer. The inclusion criteria

were the following: English language, full text articles,

original articles (retrospective or prospective), meta-

analysis. We considered studies that evaluated FDG-PET

imaging for mediastinal lymph node staging and provided

enough data to calculate sensitivity and specificity on a

per patient analysis and on a per nodal station analysis.

We analyzed the number of patients and the number of

lymph node stations sampled, sensitivity, specificity,

positive and negative predictive value and accuracy of

PET. The articles on the accuracy of PET(/CT) in the

restaging after induction radio or chemotherapy were

excluded.

3. RESULTS

A total of 1972 nodal stations were sampled in 305 pa-

tients (6.5 stations per patient): 1421 mediastinal, 287

hilar and 264 intrapulmonary nodal stations; the mean

number of lymph node dissected was 30 ± 13 per patient.

At the pathological analysis 214 patients had no lymph

node metastases, 40 patients showed an N1 disease, 50 an

N2 disease and only one an N3 disease. Integrated

PET/CT correctly staged 247 out of 305 patients; N1

disease was correctly detected by PET/CT in 34 of 40

patients (85.0%) with pathologically proven disease.

N2/N3 disease was correctly determined by PET/CT in

25 of 51 patients (49.0%) with positive results on his-

tologic analysis. PET/CT understaged 32 patients

(10.5%). The causative factors for understaging were

subcentimeter metastatic deposits in 22 patients and

PET/CT inability to distinguish between large central

tumors and adjacent mediastinal or pulmonary lymph

nodes in 10 patients. PET/CT overstaged 26 patients

(8.5%). Overstaging was due to inflammatory conditions

(n=16) and silicoanthracosis (n=10). Integrated PET/CT

showed an overall sensitivity of 64.8%, a specificity of

87.9%, a positive predictive value of 69.4%, a negative

predictive value of 85.5%, and an accuracy of 81.0% for

the detection of intrathoracic nodal metastases on a per-

patient basis (Table 2).

Out of 1.972 nodal stations histologically evaluated,

143 proved to be positive for malignancy. PET/CT cor-

rectly identified 89 metastatic lymph node stations.

E. Pelosi et al. / HEALTH 2 (2010) 204-210

207

Ta ble 2. Contingency table for PET/CT in identifying intra-

thoracic lymph node (LN) involvement.

Intrathoracic LN

involvement (+)

Intrathoracic LN

involvement (-) Tot al

PET/CT (+) 59 26 85

PET/CT (-) 32 188 220

Total 91 214 305

False negative results were obtained in 54 nodal stations,

and false positive results in 39.

The overall sensitivity, specificity, positive and nega-

tive predictive values, and accuracy of PET/CT for de-

tecting intrathoracic lymph node involvement were

62.2%, 97.9%, 69.5%, 97.1%, and 95.3%, respectively,

on a per-nodal station basis (Table 3). The most common

lymph node station for hidden metastatic involvement

was at the subcarinal level (12 out of 54 [22.2%]) fol-

lowed by the right upper and lower paratracheal and the

hilar levels.

The short-axis diameter of the 143 pathologically

proven metastatic lymph node stations ranged from 3 to

37.5 mm, with a mean value of 10.9 ± 4.8 mm. Of these

143 lymph node stations, 66 (46.1%) were less than 10

mm in short-axis diameter. The size of the 89 lymph node

stations that yielded true positive results on PET/CT

ranged from 6.5 to 37.5 mm, with a mean value of 12.9 ±

4.8 mm. Twenty (22.5%) of 89 true positive lymph node

stations at PET/CT were less than 10 mm in short-axis

diameter. The size of the 54 lymph node stations that

yielded false negative results on PET/CT ranged from

3 to 15 mm, with a mean value of 7.6 ± 2.0 mm. Forty

four (81.5%) of 54 false negative lymph node stations

were less than 10 mm in short-axis diameter. As a matter

of fact, PET/CT was successful in identifying 67 of 77

(87.0%) metastatic lymph node stations measuring ≥ 10

mm in short-axis diameter, and 22 of 66 (33.3%) metas-

tatic lymph node stations measuring < 10 mm in

short-axis diameter (p<0.001). The incidence of false

negative lymph node metastases at PET/CT was higher in

patients with adenocarcinoma (42 out of 54).

In order to evaluate PET/CT diagnostic efficacy in the

detection of mediastinal nodal metastasis a per-patient

analysis was performed in the subset of cases presenting

N0 and N1 disease. This analysis was performed due to

the central importance of mediastinal lymph node in-

volvement for therapeutic decision making in patients

with localized NSCLC. The results are presented in Table

4 (sensitivity 49.0%, specificity 96.1%, accuracy

88.2%).Patients with adenocarcinoma showed signifi-

cantly higher mediastinal nodal metastatic rates than

those with other NSCLC types. Twenty two (84.6%) of

the 26 false negative N2/N3 disease interpretations oc-

curred in patients with adenocarcinoma.

In the same subset, a per-nodal-station analysis was

performed. Sensitivity, specificity, positive and negative

Table 3. Per-nodal station diagnostic efficacy of PET/CT.

Nodal

station #

Sensitivity

Specificity

PPV

NPV

Accuracy

1 - 100 - 98.1 98.1

2 20.0 98.8 50.0 95.5 94.4

3 - 100 - 90.5 90.5

4 52.6 97.6 71.4 94.8 93.0

5 81.8 97.3 75.0 98.2 95.9

6 62.5 98.3 71.4 97.4 96.0

7 36.8 98.6 63.6 95.8 94.6

8 50 100 100 98.9 98.9

9 - 99.6 - 98.8 98.4

(n=1380) 44.7 98.8 68.0 96.8 95.8

10 86.0 91.8 64.9 97.4 90.9

11-12 78.3 98.8 85.7 97.9 97.0

(n=551) 83.3 95.3 70.5 97.7 93.8

(n=41) - 100 - 97.6 97.6

# Lymph node stations were localised according to the classification

scheme of Mountain and Dresler [18]

PPV: positive predictive value; NPV: negative predictive value

predictive values and accuracy resulted: 44.2%, 98.8%,

68%, 96.9% and 95.8%, respectively.

In Table 5 [20-29], we compared our results with the

data showed in the literature analyzing the number of

patients, the number of dissected lymph node stations and

the sensitivity, specificity, positive and negative predic-

tive values and accuracy, on a per-patient and on a per-

nodal station analysis (when possible).

4. DISCUSSION

Since the 1960’s, cervical mediastinoscopy has been

extensively performed to stage the mediastinum in pa-

tients potential candidates for lung resection. Specificity

and false positive rates of mediastinoscopy can be as-

sumed to be 100% and 0%, respectively, while in a review

of more than 6.500 patients, the average sensitivity was

approximately 80% and the average false negative rate

was nearly 10% [30]. False negative results mainly occur in

lymph node stations not attainable through media-stinoscopy;

moreover, the accuracy of the technique is surgeon dependent

[31]. In recent years, transbronchial [32] and transesophageal

[33] ultrasound-guided needle biopsy techniques have provided

Table 4. Contingency table for PET/CT in identifying medi-

astinal lymph node (LN) involvement.

Mediastinal LN

involvement (+)

Mediastinal LN

involvement (-) Tot al

PET/CT (+) 25 10 35

PET/CT (-) 26 244 270

Total 51 254 305

E. Pelosi et al. / HEALTH 2 (2010) 204-210

208

a valuable adjunct for the evaluation of mediastinoscopic

“blind spots”. Apart from the fact that mediastinoscopy is

unable to completely stage the mediastinum, it also pre-

sents several shortcomings which include invasiveness,

risk of morbidity and mortality, and costs. As a conse-

quence, different imaging techniques have been proposed

as a guide to enable the most efficient use of medi-

astinoscopy.

Several studies clearly demonstrate the key role of PET

with FDG in the preoperative staging of patients with

NSCLC, due to its ability in detecting local and distant

metastases. In a recent meta-analysis FDG-PET resulted

more accurate than CECT in identifying mediastinal

nodal metastases [29]. The average sensitivity and speci-

ficity of CECT were 61% and 79%, respectively, while

85% and 90%, respectively for PET [34]. With the in-

troduction of PET/CT, as shown for the first time by

Lardinois et al., the accuracy in detecting local and dis-

tant metastases has even increased [7,27]. However, fur-

ther studies have not confirmed these preliminary results

(see Table 5 [12-17,20]). Therefore at present the sensi-

tivity and accuracy of PET/CT in the mediastinal lymph

node staging are considered too low to avoid further

invasive studies [28], and the role of PET and medi-

astinoscopy is under discussion.

Our study confirms the limited ability of integrated

PET/CT in identifying the actual intrathoracic lymph

node stage of patients with potentially resectable NSCLC.

All the performance characteristics of integrated PET/CT

turned out to be below the threshold of 95%, only where

the test could replace invasive staging procedures [17]. In

fact, we had 10.5% of false negative and 8.5% of false

positive cases. A high rate of false negative patients for

N2 disease in the adenocarcinoma group was identified,

as with other authors [35].

Observing the available literature, we can notice a

certain heterogeneity in the methods: some authors did

not perform a per patient and a per nodal station analysis,

and/or did not report the number of the sampled stations,

and/or the average number of the dissected lymph nodes.

Anthoc et al. reported that integrated PET/CT was

more accurate than CECT in the overall tumor staging but

also in identifying nodal metastases of patients with

non-small lung cancer [8]. The sensitivity, specificity and

accuracy were 89%, 94% and 93%, respectively. The

authors explain that the advantage of integrated PET/CT

is due to the exact spatial localization of the abnormal

uptakes in the mediastinum and suggest that a mediasti-

num negative for nodal metastasis at the PET/CT study

doesn’t require any further diagnostic surgical procedure

such as mediastinoscopy.

On the contrary, Yi et al. in their study analyzed 453

mediastinal nodal stations and demonstrated that the

PET/CT sensitivity is not high enough in identifying

metastases. These authors stressed the need to continue to

perform mediastinoscopy also in stage T1, N0 patients,

despite its high cost and associated morbidity and mor-

tality. Curiously, in this study PET/CT showed very high

specificity levels leading the authors to conclude that

mediastinoscopy should not be performed in patients with

a positive PET/CT [15].

Turkmen in 2007 reported the higher accuracy of

PET/CT over CT and PET alone in detecting mediastinal

nodal metastases, but the accuracy was not enough to

replace the mediastinoscopy as the gold standard: on 59

patients he reported a sensitivity of 76% and an accuracy

of 80%. He pointed out that the causes of PET false

negative cases were micro-metastases, necrosis and PET

limited spatial resolution [36].

De Langen showed that in PET negative lymph nodes

of 10 to 15 mm the probability of discovering nodal me-

tastases is too low to justify mediastinoscopy. In con-

clusion he stressed the necessity to identify the subset of

patients with a negative PET/CT, that could benefit from

a mediastinoscopy [25-37].

Lee et al. reported that PET/CT did not improve the

overall accuracy of mediastinal staging due to the fact

that, the improvement in technology, reduced the number

Table 5. Integrated PET/CT sensibility and specificity.

Characteristic Patients (No.) LN stations (No.)Sensibility (%)Specificity (%)PPV (%) NPV (%) Accuracy (%)

Our series 305 6.5 64.8 87.9 69.4 85.5 81

Melek et al. [20] 170 4.04 74 73 55 87 74

Perigaud et al. [21] 51 3.8 40 85 40 85 -

Port et al. [22] 64 - 45 89 - - -

Nomori et al. [23] 80 - 78 98 74 98 97

Cerfolio et al. [9]* 129 N/A 69 94 49 99 96

Cerfolio et al. [24] 400 - 71 77 44 91 76

Gupta et al. [25] 77 - 87 91 72 97 82

Poncelet et al. [26] 64 - 67 85 43 94 82

Pieterman et al.* [27] 102 - 91 86 95 74 87

Kim et al. [28] 150 3.8 47 100 100 87 88

Yi et al. [29] 143 3.2 56 100 100 88 90

E. Pelosi et al. / HEALTH 2 (2010) 204-210

209

of false negative patients while it increased that of the

false positive results [16].

Cerfolio showed that the use of PET/CT in the selec-

tion of patients candidates for surgery is useful; however,

it concluded that PET/CT is not accurate enough to re-

place nodal biopsies to stage the mediastinum [9].

In the present paper, we performed both the per patient

and the per nodal station analysis in order to better un-

derstand the overall sensitivity, specificity and accuracy

of PET/CT. It is our opinion that the accuracy of PET/CT

could be more precisely evaluated by analyzing the re-

sults in all the nodal stations (it should be pointed out, that

even if other authors, including Cerfolio, stressed the

relevance of the nodal station analysis, only few authors

used it [24]). In our study, at the per nodal station analysis,

the most common location for mediastinal lymph node

metastases was the subcarinal station, followed by the

lower paratracheal station. In agreement with the findings

reported by Cerfolio and co-workers [9], PET/CT showed

the highest incidence of both false positive and false

negative results at this level, which underline the need for

a thorough lymph node dissection at the subcarinal station.

Moreover, previous or concomitant inflammatory and

infectious conditions are mainly responsible for false

positive results at PET/CT. Due to this reason and in

accordance with other authors we agree on the necessity

to confirm PET positive findings with a biopsy [16-24].

Therefore, even if PET/CT is more accurate than CT and

other preoperative exams in detecting distant and local

metastases, both the published literature and our experi-

ence underline the continued need for tissue confirmation

of a positive PET/CT result.

On the other hand, the spatial resolution of PET/CT

remains inadequate to rule out sub-centimetre lymph

node metastases and this was clearly confirmed by Peri-

gaud et al. [21]. Furthermore, Al Sarraf and coll. [38] in

their series showed that the rate of false negative patients

resulted 16% (25 out of 153). In this study, the elements

significantly related with occult N2 disease were: central

location of the tumor, right upper lobe cancer, and N1

uptake at PET/CT. Melek in his paper reported an higher

false negative rate in patients with adenocarcinoma [20],

which is confirmed by our results.

In conclusion our data show that integrated PET/CT is

more accurate than CECT in detecting nodal metastases;

however, its accuracy is below 95% and therefore, not

sufficient enough to substitute mediastinoscopy in all

patients. Therefore, in patient cohorts with a higher like-

lihood of mediastinal lymph node involvement, a nega-

tive PET scan does not obviate the need of invasive

staging procedures [31-35]. New randomized trials will

be necessary to identify these cohorts of patients. Vice-

versa, in patients with PET positive mediastinal lymph

nodes, the mediastinoscopy is mandatory in order to

avoid false positive results.

REFERENCES

[1] Jemal, A., Siegel, R., Ward, E., Murray, T., Xu, J., Smigal,

C., and Thun, M.J. (2008) Pattern and predictors of occult

mediastinal lymph node involvement in non-small cell

lung cancer patients with negative mediastinal uptake on

positron emission tomography. European Journal of

Cardio-thoracic Surgery, 56, 106-130.

[2] Yang, P., Allen, M.S., and Aubry, M.C. (2005) Clinical

features of 5,628 primary lung cancer patients: experience

at Mayo clinic from 1997 to 2003. Chest, 128, 452-462.

[3] Dwamena, B.A., Sonnad, S.S., Angobaldo, J.O., and Wahl,

R.L. (1999) Metastases from non-small cell lung cancer:

Mediastinal staging in the 1990s—meta-analytic compa-

rison of PET and CT. Radiology, 213, 530-536.

[4] Toloza, E.M., Harpole, L., and McCrory, D.C. (2003)

Noninvasive staging of non-small cell lung cancer: a re-

view of the current evidence. Chest, 123, 137S-146S.

[5] Gould, M.K., Kuschner, W.G., Rydzak, C.E., Maclean,

C.C., Demas, A.N., Shigemitsu, H., Chan, J.K., and

Owens, D.K. (2003) Test performance of positron emis-

sion tomography and computed tomography for medi-

astinal staging in patients with non-small-cell lung cancer.

A meta-analysis. Annals of Internal Medicine, 139, 879-

892.

[6] Silvestri, G.A., Gould, M.K., Margolis, M.L., Tanoue, L.T.,

McCrory, D., Toloza, E., and Detterbeck, F., American

College of Chest Physicians. (2007) Noninvasive staging

of non-small cell lung cancer: ACCP evidence-based

clinical practice guidelines (2nd edition). Chest, 132,

178S-201S.

[7] Lardinois, D., Weder, W., Hany, T.F., Kamel, E.M.,

Korom, S., Seifert, B., Von Schulthess, G.K., and Steinert,

HC. (2003) Staging of non-small-cell lung cancer with

integrated positron-emission tomography and computed

tomography. New England Journal of Medicine, 348,

2500-2507.

[8] Antoch, G., Stattaus, J., Nemat, A.T., Marnitz, S., Beyer, T.,

Kuehl, H., Bockisch, A., Debatin, J.F., and Freudenberg,

L.S. (2003) Non-small cell lung cancer: dual-modality

PET/CT in preoperative staging. Radiology, 229, 526-533.

[9] Cerfolio, R.J., Ojha, B., Bryant, A.S., Raghuveer, V.,

Mountz, J.M., and Bartolucci, A.A. (2004) The accuracy

of integrated PET/CT compared with dedicated PET alone

for the staging of patients with non-small cell lung cancer.

Annals of Thoracic Surgery, 78, 1017-1023.

[10] Halpern, B.S., Schiepers, C., Weber, W.A., Crawford, T.L.,

Fueger, B.J., Phelps, M.E., and Czernin, J. (2005) Pre-

surgical staging of non-small cell lung cancer. Positron

emission tomography, integrated positron emission to-

mography/CT, and software image fusion. Chest, 128,

2289-2297.

[11] Martini, N., Kris, M.G., and Ginsberg, R.J. (1997) The

role of multimodality therapy in locoregional non-small

cell lung cancer . Surgical Oncology Clinics of North

America, 111, 1718-1723.

[12] Cerfolio, R.J., Bryant, A.S., Ojha, B., and Eloubeidi, M.

(2005) Improving the inaccuracies of clinical staging of

patients with NSCLC: A prospective trial. Annals of

Thoracic Surgery, 80, 1207-1214.

[13] Shim, S.S., Lee, K.S., Kim, B-T, Chung, M.J., Lee, E.J.,

E. Pelosi et al. / HEALTH 2 (2010) 204-210

210

Han, J., Choi, J.Y., Kwon, O.J., Shim, Y.M., and Kim, S.

(2005) Non-small cell lung cancer: prospective compari-

son of integrated FDG PET/CT and CT alone for preop-

erative staging. Radiology, 236, 1011-1019.

[14] Kim, B-T, Lee, K.S., Shim, S.S., Choi, J.Y., Kwon, O.J.,

Kim, H., Shim, Y.M., Kim, J., and Kim, S. (2006) Stage

T1 non-small cell lung cancer: preoperative mediastinal

nodal staging with integrated FDG PET/CT — a pro-

spective study. Radiology, 241, 501-509.

[15] Yi, C.A., Lee, K.S., Kim, B-T, Shim, S.S., Chung, M.J.,

Sung, Y.M., and Jeong, S.Y. (2007) Efficacy of helical

dynamic CT versus integrated PET/CT for detection of

mediastinal nodal metastasis in non-small cell lung cancer.

Australian Journalism Review, 188, 318-325.

[16] Lee, B.E., Von Haag, D., Lown, T., Lau, D., Calhoun, R.,

and Follette, D. (2007) Advances in positron emission

tomography technology have increased the need for sur-

gical staging in non-small cell lung cancer. Journal of

Thoracic and Cardiovascular Surgery, 133, 746-752.

[17] Tournoy, K.G., Maddens, S., Gosselin, R., Van Maele, G.,

Van Meerbeeck, J.P., and Kelles, A. (2007) Integrated

FDG-PET/CT does not make invasive staging of the in-

trathoracic lymph nodes in non-small cell lung cancer

redundant: a prospective study. Thorax, 62, 696-701.

[18] Hammound, Z.T., Anderson, R.C., Meyers, B.F., Guthrie,

T.J., Roper, C.L., Cooper, J.D., and Patterson, G.A. (1999)

The current role of mediastinoscopy in the evaluation of

thoracic disease. Journal of Thoracic and Cardiovascular

Surgery, 118, 894-899.

[19] Mountain, C.F., and Dresler, C.M. (1997) Regional lymph

node classification for lung cancer staging. Chest, 111,

1718-1723.

[20] Melek, H., Gunluoglu, M.Z., Demir, A., Akin, H., Olcmen,

A., and Dincer, S.I. (2008) Role of positron emission to-

mography in mediastinal lymphatic staging of non-small

cell lung cancer. European Journal of Cartiothoracic

Surgery, 33, 294-299.

[21] Perigaud, C., Bridji, B., Roussel, J.C., Sagan, C., Mugniot,

A., Duveau, D., Baron, O., and Despins, P. (2009) Pro-

spective preoperative mediastinal lymph node staging by

positron emission tomography computerized tomography

in patients with non-small-cell lung cancer. European

Journal of Cartiothoracic Surgery, 36,731-736.

[22] Port, J.L. (2005) Positron emission tomographic scanning

in the diagnosis and staging of non-small cell lung cancer

2 cm in size or less. Journal of Thoracic and Cardiovas-

cular Surgery, 130, 611-615

[23] Nomori, H., Watanabe, K., Ohtsuka, T., Naruke, T., Sue-

masu, K., Kobayashi, T., and Uno, K. (2004) Fluorine

18-tagged fluorodeoxyglucose positron emission tomo-

graphy scanning to predict lymph node metastasis, inva-

siviness, or both in clinical T1 N0 M0 lung adenocarci-

noma. Journal of Thoracic and Cardiovascular Surgery,

128, 396-401.

[24] Cerfolio, R.J., Ojha, B., Briant, A.S., Brass, C.S., Barta-

lucci, A.A., and Mountz, J.M. (2003) The role of FDG-

PET scan in staging patients with non small cell carci-

noma. Annals of Thoracic Surgery, 76, 861-866

[25] Gupta, N.C. (2001) Mediastinal lymph node sampling

following positron emission tomography with fluoride-

oxyglucose imagin gin lung cancer staging. Chest, 120,

521-527.

[26] Poncelet, A.J., Lonneux, M., Coche, E., Weynand, B., and

Moirhomme, Ph. (2001) FDG-PET scan enhances but

does not replace surgical staging in non-small cell lung

carcinoma. European Journal of Cartiothoracic Surgery,

20, 468-475.

[27] Pieterman, R.M. (2000) Preoperative staging of non-small

cell lung cancer with positron-emission tomography. New

England Journal of Medicine, 343, 254-261.

[28] Kim, B-T, Lee, K.S., Shim, S.S., Choi, J.Y., Kwon, O.J.,

Kim, H., Shim, Y.M., Kim, J., and Kim, S. (2006) Stage

T1 non-small cell lung cancer: preoperative mediastinal

nodal staging with integrated FDG PET/CT —a prospec-

tive study. Radiology, 241, 501-509.

[29] Yi, C.A., Lee, K.S., Kim, B-T, Shim, S.S., Chung, M.J.,

Sung, Y.M., and Jeong, S.Y. (2007) Efficacy of helical

dynamic CT versus integrated PET/CT for detection of

mediastinal nodal metastasis in non-small cell lung cancer.

Australian Journalism Review, 188, 318-325.

[30] Detterbeck, F.C., Jantz, M.A., Wallace, M., Vansteenkiste,

J., Silvestri, G.A. (2007) Invasive mediastinal staging of

lung cancer: ACCP evidence-based clinical practice

guidelines (2nd edition). Chest, 132, 202S-220S.

[31] Smulders, S.A., Smeenk, FWJW, Janssen-Heijnen, MLG,

Wielders, PLML, de Munck, DRAJ, and Postmus, P.E.

(2005) Surgical mediastinal staging in daily practice.

Lung Cancer, 47, 243-251.

[32] Herth, F.J., Eberhardt, R., Vilmann, P., Krasnik, M., and

Ernst, A. (2006) Real-time endobronchial ultrasound

guided transbronchial needle aspiration for sampling me-

diastinal lymph nodes. Thorax, 61, 795-798.

[33] Annema, J.T., Versteegh, M.I., Veseliç, M., Welker, L.,

Mauad, T., Sont, J.K., Willems, L.N., and Rabe, K.F.

(2005) Endoscopic ultrasound added to mediastinoscopy

for preoperative staging of patients with lung cancer.

Journal of Clinical Investigation, 294, 931-936.

[34] Behzadi, A., Ung, Y., Lowe, V., and Deschamps, C. (2008)

The role of positron emission tomography in the man-

agement of non-small cell lung cancer. Canadian Journal

of Cardiology, 3, 235-242.

[35] Al-Sarraf, N., Aziz, R., Doddakula, K., Gately, K., Wilson,

L., McGovern, E., and Young, V. (2006) Factors causing

inaccurate staging of mediastinal nodal involvement in

non-small cell lung cancer patients staged by positron

emission tomography. Interactive CardioVascular and

Thoracic Surgery.

[36] Turkman, C., Sommezoglu, K., and Toker, A. (2007) The

additional value of FDG PET imaging for distinguishing

N0 or N1 from N2 stage in preoperative staging of non-small

cell lung cancer in region where the prevalence of inflam-

matory lung disease is high. Nuclear Medicine, 32, 607-612.

[37] De Langen, A.J., Raijmakers, P., and Riphagen, I. (2006)

The size of mediastinal lymph nodes and its relation with

metastatic involvement: A meta-analysis. European Jour-

nal of Cartiothoracic Surgery, 29, 26-29.

[38] Al-Sarraf, N., Aziz, R., Gately, K., Lucey, J., Wilson, L.,

McGovern, E., and Young, V. (2008) Pattern and predic-

tors of occult mediastinal lymph node involvement in

non-small cell lung cancer patients with negative medi-

astinal uptake on positron emission tomography. Euro-

pean Journal of Cartiothoracic Surgery, 33,104-109.