Difference in interpretation of computed tomography pulmonary angiography diagnosis of subsegmental thrombosis in patients with suspected pulmonary embolism

Abstract

The introduction of computed tomography pulmonary angiography (CTPA) as a diagnostic modality to diagnose pulmonary embolism (PE) has led to a significant increase in the incidence of PE diagnosis. Multidetector CTPA has improved the sensitivity for diagnosis of PE, allowing better visualization of segmental and subsegmental pulmonary arteries [1Rathbun S.W. Raskob G. Whitsett T.L. Sensitivity and specificity of helical computed tomography in the diagnosis of pulmonary embolism: a systematic review.Ann Intern Med. 2000; 132: 227-32Crossref PubMed Google Scholar, 2Mullins M.D. Becker D.M. Hagspiel K.D. Philbrick J.T. The role of spiral volumetric computed tomography in the diagnosis of pulmonary embolism.Arch Intern Med. 2000; 160: 293-8Crossref PubMed Google Scholar, 3Le Gal G. Righini M. Parent F. van Strijen M. Couturaud F. Diagnosis and management of subsegmental pulmonary embolism.J Thromb Haemost. 2006; 4: 724-31Crossref PubMed Scopus (142) Google Scholar, 4Ghaye B. Szapiro D. Mastora I. Delannoy V. Duhamel A. Remy J. Remy‐Jardin M. Peripheral pulmonary arteries: how far in the lung does multi‐detector row spiral CT allow analysis?.Radiology. 2001; 219: 629-36Crossref PubMed Google Scholar]. Concomitantly, the proportion of patients with suspected PE in whom isolated subsegmental pulmonary embolism (SSPE) is reported has increased [5Carrier M. Righini M. Wells P.S. Perrier A. Anderson D.R. Rodger M.A. Pleasance S. Le Gal G. Subsegmental pulmonary embolism diagnosed by computed tomography: incidence and clinical implications. A systematic review and meta‐analysis of the management outcome studies.J Thromb Haemost. 2010; 8: 1716-22Crossref PubMed Scopus (0) Google Scholar]. Although the incidence of PE diagnosis has increased within the last few years, its case fatality rate and the mortality rates have remained unchanged, suggesting that the additional cases of PE might be associated with a lower severity of illness [6Wiener R.S. Schwartz L.M. Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis.Arch Intern Med. 2011; 171: 831-7Crossref PubMed Scopus (516) Google Scholar, 7DeMonaco N.A. Dang Q. Kapoor W.N. Ragni M.V. Pulmonary embolism incidence is increasing with use of spiral computed tomography.Am J Med. 2008; 121: 611-7Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar]. The management of PE isolated to the subsegmental pulmonary arteries (single or multiple vessels) without a concomitant deep vein thrombosis (DVT) is controversial among thrombosis experts and clinical equipoise exists regarding the use of anticoagulant therapy in the management of those isolated SSPEs [8Carrier M. Kimpton M. Le Gal G. Kahn S.R. Kovacs M.J. Wells P.S. Anderson D.R. Rodger M.A. The management of a sub‐segmental pulmonary embolism: a cross‐sectional survey of Canadian thrombosis physicians.J Thromb Haemost. 2011; 9: 1412-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. Given how important it is for clinicians to have an accurate diagnosis in order to decide on the risk‐benefit ratio of anticoagulant therapy to treat SSPE, we sought to determine the proportion of agreement between radiologists on the diagnosis of isolated SSPE in patients presenting with suspected PE, and to report the management and 3‐month outcome of patients who had been diagnosed with an isolated SSPE. We conducted a retrospective cohort study of consecutive patients with suspected acute PE who underwent CTPA at the Ottawa Hospital between January 2007 and December 2008. Computed tomography pulmonary angiography was performed using either GE LightSpeed Plus (4‐ and 16‐detectors), GE LightSpeed VCT (64‐detectors) or Toshiba Aquilion One (320‐detectors) CT devices. SSPE was defined as any number of pulmonary artery filling defects isolated to the subsegmental pulmonary arteries detected on CTPA. All included patients were followed for 3 months. The CTPA images of all patients with an index diagnosis of SSPE were reviewed by an experienced thoracic radiologist (RP). The examiner was blinded to the diagnosis and original filling defect location. The reviewing radiologist was also blinded to any other diagnostic test results (compressive lower extremity ultrasonography [US], etc.). The reviewing radiologist independently recorded findings on standardized forms. Normal CTPA (n = 4) studies and CTPA studies with index segmental filling defects (n = 2) interpretation were also included to ensure control and blindness of the reviewing radiologist. Disagreements were reviewed by a second experienced thoracic radiologist (CD). All clinical outcomes captured during follow‐up were reviewed independently by two investigators (MK and MC). Disagreements on information were resolved by consensus or by retrieving further information from other medical records. The primary outcome was symptomatic objectively confirmed recurrent venous thromboembolism (VTE) during the 3‐month follow‐up period. Secondary outcomes included major bleeding episodes and overall mortality. Recurrent PE was defined as a new arterial filling defect on CTPA or high probability V/Q‐scan, and recurrent DVT was defined as a non‐compressible venous segment on an ultrasound of the extremities. Major bleeding events were defined as per the Internal Society of Thrombosis and Haemostasis definition [9Schulman S. Kearon C. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non‐surgical patients.J Thromb Haemost. 2005; 3: 692-4Crossref PubMed Scopus (3059) Google Scholar]. All study outcomes were adjudicated. Descriptive statistics were used to describe the patient characteristics and outcomes. We computed the proportion of cases, with its associated 95% confidence intervals (CI), in whom the second interpretation by the thoracic radiologist was concordant with the initial diagnosis of isolated SSPE. Agreement was determined using the most proximal level of filling defect detected on CTPA images. Analyses were performed using StatsDirect software version 2.7.3 (StatsDirect Ltd, Altrincham, UK). Over the study period, 4410 CTPAs were carried out for suspected symptomatic PE at The Ottawa Hospital. Out of 724 (16.4%) patients who had a CTPA positive for PE, 70 (9.6%) patients had a diagnosis of isolated SSPE. The median age was 64 (range 15–92) and 55% (39/70) were female. Twenty‐six per cent had a prior history of VTE. Forty‐five per cent of patients had cancer. The thoracic radiologist reviewed 76 CTPA studies (SSPE, n = 70; segmental PE, n = 2; normal, n = 4) (See Table 1). Ninety per cent (69/76) of studies were described as having adequate opacification and 18% (14/76) had significant breathing artifact by the reviewing thoracic radiologist.Table 1Index and thoracic radiologist’s interpretations of the most proximal filling defect detected on CTPAIndex interpretationLobarSegmentalSubsegmentalNormalTotalThoracic radiologist’s interpretation00000 Lobar0226028 Segmental Subsegmental0036036 Normal008412Total0270476CTPA, computed tomographic pulmonary angiography. Open table in a new tab CTPA, computed tomographic pulmonary angiography. The reviewing radiologist agreed with the initial diagnosis of isolated SSPE in 51% (36/70; 95% CI, 39–64%) of the cases. Of the 36 CTPA examinations with an agreement on an isolated SSPE diagnosis, 13 and 23 patients had an index diagnosis of single and multiple isolated SSPE, respectively. The reviewing radiologist agreed with the number of filling defects (i.e. single vs. multiple isolated SSPE) in 83% (30/36; 95% CI, 67–94%) of the cases. A total of 11% of the included CTPA examinations with an index diagnosis of SSPE were re‐interpreted by the thoracic radiologist to be without any evidence of PE. However, 37% (26/70) of patients with an index diagnosis of SSPE were re‐interpreted to have segmental defects (21 single and five multiple segmental defects). Out of the 70 patients with an index SSPE diagnosis, 52 patients (74%) received anticoagulation. One patient (2%) who received anticoagulation suffered a major bleeding episode. None of the 18 patients who did not receive any anticoagulation (14 single and four multiple isolated SSPE) suffered recurrent VTE (PE or DVT) during the 3‐month follow‐up period (0%; 95% CI, 0–18.5%). Thirty‐three per cent (23/70) of patients died during the follow‐up period, most of them from cancer progression. No death was adjudicated as being due to PE. A total of 53 (76%) patients had bilateral compressive US of the lower extremities and seven (10%) patients were diagnosed with a proximal DVT. All patients with DVT were managed with anticoagulation. The seven patients with DVT had CTPA examinations re‐interpreted as multiple (n = 5) or single (n = 2) segmental PE. There are significant differences in the interpretations of a diagnosis of isolated SSPE among radiologists. This uncertainty around the diagnosis could potentially lead to inappropriate use of anticoagulation and related complications (major bleeding episodes or recurrent VTE). The diagnosis of isolated SSPE on CTPA should be reviewed by an experienced thoracic radiologist before physicians make any clinical decision on anticoagulant therapy. Our results are consistent with previously published literature. A very good interobserver agreement was recently reported for the diagnosis of massive PE using CTPA (overall κ = 0.82; 95% CI, 0.68–0.95) [10Costantino G. Norsa A.H. Amadori R. Ippolito S. Resta F. Bianco R. Casazza G. Biagiotti S. Rusconi A.M. Montano N. Interobserver agreement in the interpretation of computed tomography in acute pulmonary embolism.Am J Emerg Med. 2009; 27: 1109-11Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar]. However, the concordance was lower with regard to segmental and subsegmental PE (κ = 0.47; 95% CI, 0.16–0.84) [10Costantino G. Norsa A.H. Amadori R. Ippolito S. Resta F. Bianco R. Casazza G. Biagiotti S. Rusconi A.M. Montano N. Interobserver agreement in the interpretation of computed tomography in acute pulmonary embolism.Am J Emerg Med. 2009; 27: 1109-11Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar]. Similarly, another interobserver agreement study reported a κ of 0.38 (95% CI: 0.0–0.89) for SSPE among radiologists with varied levels of experience [11Ghanima W. Nielssen B.E. Holmen L.O. Witwit A. Al‐Ashtari A. Sandset P.M. Multidetector computed tomography (MDCT) in the diagnosis of pulmonary embolism: interobserver agreement among radiologists with varied levels of experience.Acta Radiol. 2007; 48: 165-70Crossref PubMed Scopus (51) Google Scholar]. However, this study is limited due to the small number of SSPEs included with the analyses. The clinical importance of an SSPE diagnosis is unknown [3Le Gal G. Righini M. Parent F. van Strijen M. Couturaud F. Diagnosis and management of subsegmental pulmonary embolism.J Thromb Haemost. 2006; 4: 724-31Crossref PubMed Scopus (142) Google Scholar, 12Eyer B.A. Goodman L.R. Washington L. Clinicians’ response to radiologists’ reports of isolated subsegmental pulmonary embolism or inconclusive interpretation of pulmonary embolism using MDCT.AJR Am J Roentgenol. 2005; 184: 623-8Crossref PubMed Google Scholar, 13Goodman L.R. Small pulmonary emboli: what do we know?.Radiology. 2005; 234: 654-8Crossref PubMed Scopus (130) Google Scholar]. The increased incidence of SSPE with CTPA seems to be associated with a lower severity of illness in the CTPA era [5Carrier M. Righini M. Wells P.S. Perrier A. Anderson D.R. Rodger M.A. Pleasance S. Le Gal G. Subsegmental pulmonary embolism diagnosed by computed tomography: incidence and clinical implications. A systematic review and meta‐analysis of the management outcome studies.J Thromb Haemost. 2010; 8: 1716-22Crossref PubMed Scopus (0) Google Scholar, 6Wiener R.S. Schwartz L.M. Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis.Arch Intern Med. 2011; 171: 831-7Crossref PubMed Scopus (516) Google Scholar, 7DeMonaco N.A. Dang Q. Kapoor W.N. Ragni M.V. Pulmonary embolism incidence is increasing with use of spiral computed tomography.Am J Med. 2008; 121: 611-7Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar, 14Auer R.C. Schulman A.R. Tuorto S. Gonen M. Gonsalves J. Schwartz L. Ginsberg M.S. Fong Y. Use of helical CT is associated with an increased incidence of postoperative pulmonary emboli in cancer patients with no change in the number of fatal pulmonary emboli.J Am Coll Surg. 2009; 208: 871-8Crossref PubMed Scopus (47) Google Scholar]. A recently completed randomized controlled trial comparing the utility of CTPA with ventilation/perfusion scanning for the management of patients with suspected PE has shown that CTPA resulted in a significantly greater number of VTE diagnoses than did ventilation/perfusion scans; hence, more patients diagnosed by CTPA were treated with anticoagulants [15Anderson D.R. Kahn S.R. Rodger M.A. Kovacs M.J. Morris T. Hirsch A. Lang E. Stiell I. Kovacs G. Dreyer J. Dennie C. Cartier Y. Barnes D. Burton E. Pleasance S. Skedgel C. O’Rouke K. Wells P.S. Computed tomographic pulmonary angiography vs ventilation‐perfusion lung scanning in patients with suspected pulmonary embolism: a randomized controlled trial.JAMA. 2007; 298: 2743-53Crossref PubMed Google Scholar]. Despite this, the rate of VTE during the 3‐month follow‐up period was not different in untreated patients (i.e. in whom PE was excluded) who were randomized to either diagnostic strategy, suggesting that the additional cases of PE detected by CTPA were clinically unimportant. In our study, approximately 10% of patients with an index diagnosis of SSPE were found to have no evidence of PE upon re‐interpretation by an experienced thoracic radiologist. When confronted with a diagnosis of isolated SSPE, a majority of clinicians will initiate anticoagulant therapy [8Carrier M. Kimpton M. Le Gal G. Kahn S.R. Kovacs M.J. Wells P.S. Anderson D.R. Rodger M.A. The management of a sub‐segmental pulmonary embolism: a cross‐sectional survey of Canadian thrombosis physicians.J Thromb Haemost. 2011; 9: 1412-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar, 16Donato A.A. Khoche S. Santora J. Wagner B. Clinical outcomes in patients with isolated subsegmental pulmonary emboli diagnosed by multidetector CT pulmonary angiography.Thromb Res. 2010; 126: e266-70Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar]. Current guidelines recommend anticoagulating patients with unprovoked VTE for a minimum of 3 months and to consider long‐term treatment in the absence of contraindications [17Kearon C. Kahn S.R. Agnelli G. Goldhaber S.Z. Raskob G.E. Comerota A.J. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence‐based Clinical Practice Guidelines (8th edition).Chest. 2008; 133: 454-545Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. Therefore, accurate diagnosis of SSPE is important before exposing patients to potentially indefinite anticoagulant therapy and its associated bleeding risk, which has been reported to occur in 5.3% of patients with isolated SSPE receiving anticoagulants [16Donato A.A. Khoche S. Santora J. Wagner B. Clinical outcomes in patients with isolated subsegmental pulmonary emboli diagnosed by multidetector CT pulmonary angiography.Thromb Res. 2010; 126: e266-70Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar]. A previous survey study has shown that 88% and 57% of thrombosis experts felt comfortable withholding anticoagulant therapy and using an alternative management strategy (Doppler ultrasonography, ventilation/perfusion scans, etc.) in patients with unprovoked single and multiple isolated SSPE, respectively [8Carrier M. Kimpton M. Le Gal G. Kahn S.R. Kovacs M.J. Wells P.S. Anderson D.R. Rodger M.A. The management of a sub‐segmental pulmonary embolism: a cross‐sectional survey of Canadian thrombosis physicians.J Thromb Haemost. 2011; 9: 1412-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. Less than 20% felt comfortable withholding anticoagulant therapy in patients with single segmental PE [8Carrier M. Kimpton M. Le Gal G. Kahn S.R. Kovacs M.J. Wells P.S. Anderson D.R. Rodger M.A. The management of a sub‐segmental pulmonary embolism: a cross‐sectional survey of Canadian thrombosis physicians.J Thromb Haemost. 2011; 9: 1412-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. Interestingly, in our study, 37% of patients with an index SSPE diagnosis were re‐interpreted to have a more proximal filling defect (segmental). Hence, accuracy on the level (subsegmental, segmental, etc.) of the filling defect for a PE diagnosis is important and could potentially have significant clinical consequences for patient care. There is no clear consensus regarding management of patients with isolated SSPE. To date, over 50 patients diagnosed with isolated SSPE on CTPA performed for suspected PE who did not receive anticoagulation treatment have been reported in the literature [3Le Gal G. Righini M. Parent F. van Strijen M. Couturaud F. Diagnosis and management of subsegmental pulmonary embolism.J Thromb Haemost. 2006; 4: 724-31Crossref PubMed Scopus (142) Google Scholar, 16Donato A.A. Khoche S. Santora J. Wagner B. Clinical outcomes in patients with isolated subsegmental pulmonary emboli diagnosed by multidetector CT pulmonary angiography.Thromb Res. 2010; 126: e266-70Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar]. None of these patients had a recurrent fatal PE. Similarly, none of the 18 patients with confirmed isolated SSPE that were left untreated in our cohort study had a recurrent VTE within the 3‐month follow‐up period. This may suggest that withholding anticoagulant treatment in patients with isolated SSPE without DVT might be a safe management. However, prospective cohort management studies are required before withholding anticoagulant therapy becomes common practise in patients with isolated SSPE. In summary, our results indicate that significant differences exist in the interpretations of a diagnosis of isolated SSPE on CTPA among radiologists. Given the clinical equipoise that exists around the management of single and multiple isolated SSPE, CTPA examinations with isolated SSPE diagnosis should be reviewed by an experienced thoracic radiologist to ensure proper risk‐benefit ratio assessment and clinical management of patient care. The authors state that they have no conflict of interest.