0
Correspondence |

Interobserver Variability in Grading Acute Rejection After Lung TransplantationVariability in Grading Posttransplant Rejection FREE TO VIEW

Wenjun Mao, MD; Wei Xia, MD; Jingyu Chen, MD
Author and Funding Information

From the Department of Cardiothoracic Surgery (Drs Mao and Chen) and the Department of Intensive Care Unit (Dr Xia), Wuxi People’s Hospital, Nanjing Medical University.

Correspondence to: Wenjun Mao, MD, Division of Cardiothoracic Surgery, Wuxi People’s Hospital, No. 299, Qing Yang Rd, Wuxi City, Jiangsu, China 214023; e-mail: maowenjun1@126.com


Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2014;145(2):416-417. doi:10.1378/chest.13-1788
Text Size: A A A
Published online
To the Editor:

We read with great interest the article by Bhorade et al1 in CHEST (June 2013) about the interobserver agreement of grading acute rejection after lung transplantation. They indicated that the overall concordance rates for grade A and grade B biopsy specimens were 74% and 89%, respectively, and interobserver discrepancies for acute rejection were lower when pulmonary biopsies were performed earlier (≤ 6 weeks) compared with later time points. However, we would like to add more information after deeper analysis of the data and address some important concerns.

In the Bhorade et al1 study, the interobserver agreement for grade A and grade B readings were presented as the overall concordance rate, as well as that determined by treatment arm and clinical symptoms. The overall concordance rate ranged from 62% to 91%, according to the data from the tables in the article1; however, it should be noted that the interobserver agreement provided by the authors was ambiguous and requires further analysis. Therefore, we conducted κ analysis to reevaluate the concordance of interpretations for acute rejection between site pathologist and central pathologist (based on the data presented in tables in the article). The score of Cohen κ coefficients ranged from 0 to 1, where κ scores ≥ 0.75 represent fair agreement, scores < 0.4 represent poor agreement, and the scale of 0.4 to 0.75 was considered moderate agreement. The McNemar-Bowker test was performed to estimate the diagnostic differences between site pathologist and central pathologist. After thorough statistical analysis of the data from Tables 2 and 3 in the Bhorade et al1 article, we found that site pathologists were more likely to judge the acute rejection at a higher level for both grade A readings and grade B readings (P = .000 and .002, respectively), and the κ scores showed poor interobserver agreement for both grade A and grade B readings (κ = 0.276 and 0.195, respectively).

Also, as we can see from Table 1, when the effect of the immunosuppressant type and symptoms in interobserver agreement for grade A readings were reassessed, moderate agreement for the azathioprine group was achieved (κ = 0.472), whereas poor agreement for the sirolimus group, surveillance, and clinical biopsy specimens (κ < 0.4) was found, and more acute rejections at a higher level were diagnosed by site pathologists when compared with central pathologists (P < .05). Poor concordance was observed for grade B readings (κ < 0.4) (Table 2), and discrepancy in diagnostic ability existed between the site and central pathologists for the azathioprine group and surveillance biopsy specimens (P = .015 and .011, respectively); however, there was no difference regarding the diagnostic ability for the sirolimus group and clinical biopsy specimens (P = .096 and .146, respectively). Hence, it is important to recognize the differences in diagnostic capability for acute rejection between central and site pathologists in the context of the current grading system.

Table Graphic Jump Location
Table 1 —κ Analysis and McNemar-Bowker Test for Grade A Readings
Table Graphic Jump Location
Table 2 —κ Analysis and McNemar-Bowker Test for Grade B Readings

The data analysis as an addition to the article reflects that results may be confused by some factors that complicate clinical decision-making. Identification of risk factors as predictors of acute rejection after lung transplantation may facilitate the diagnosis of acute rejection and reduce morbidity. Some studies reported that patients with human leukocyte antigen (HLA) mismatches at the B and DR loci were more susceptible to acute rejection, implying that bias introduced by mismatches at the HLA-DR and HLA-B loci may affect interobserver agreement for acute rejection after transplant,2,3 which should be included in the statistical analysis. Moreover, whether the transplant type influences the interobserver agreement remains undecided.

Bhorade et al1 mentioned that the concordance rate of scheduled biopsy specimens was higher than that of clinical bronchoscopy.1 Not only the blinded analysis, but the complex state of the illness per se often complicated the specimen readings when patients presented with clinical symptoms, with greater deviations made both by the site and central pathologists. Therefore, repeated blinded readings of one slide by the same pathologist as a reliable policy is a promising approach to decrease the interobserver disagreement. In addition, although further education for pathologists to improve concordance in grading acute rejection was recommended by the authors,1 optimal modifications to the International Society for Heart and Lung Transplantation grading system for acute rejection after lung transplantation should be made with the accumulation of clinical experience resembling the evolution of TNM classification for non-small-cell lung cancer.4,5

References

Bhorade SM, Husain AN, Liao C, et al. Interobserver variability in grading transbronchial lung biopsy specimens after lung transplantation. Chest. 2013;143(6):1717-1724. [CrossRef] [PubMed]
 
Schulman LL, Weinberg AD, McGregor C, Galantowicz ME, Suciu-Foca NM, Itescu S. Mismatches at the HLA-DR and HLA-B loci are risk factors for acute rejection after lung transplantation. Am J Respir Crit Care Med. 1998;157(6 pt 1):1833-1837. [CrossRef] [PubMed]
 
Mangi AA, Mason DP, Nowicki ER, et al. Predictors of acute rejection after lung transplantation. Ann Thorac Surg. 2011;91(6):1754-1762. [CrossRef] [PubMed]
 
Detterbeck FC, Boffa DJ, Tanoue LT. The new lung cancer staging system. Chest. 2009;136(1):260-271. [CrossRef] [PubMed]
 
Saji H, Tsuboi M, Shimada Y, et al. A proposal for combination of total number and anatomical location of involved lymph nodes for nodal classification in non-small cell lung cancer. Chest. 2013;143(6):1618-1625. [CrossRef] [PubMed]
 

Figures

Tables

Table Graphic Jump Location
Table 1 —κ Analysis and McNemar-Bowker Test for Grade A Readings
Table Graphic Jump Location
Table 2 —κ Analysis and McNemar-Bowker Test for Grade B Readings

References

Bhorade SM, Husain AN, Liao C, et al. Interobserver variability in grading transbronchial lung biopsy specimens after lung transplantation. Chest. 2013;143(6):1717-1724. [CrossRef] [PubMed]
 
Schulman LL, Weinberg AD, McGregor C, Galantowicz ME, Suciu-Foca NM, Itescu S. Mismatches at the HLA-DR and HLA-B loci are risk factors for acute rejection after lung transplantation. Am J Respir Crit Care Med. 1998;157(6 pt 1):1833-1837. [CrossRef] [PubMed]
 
Mangi AA, Mason DP, Nowicki ER, et al. Predictors of acute rejection after lung transplantation. Ann Thorac Surg. 2011;91(6):1754-1762. [CrossRef] [PubMed]
 
Detterbeck FC, Boffa DJ, Tanoue LT. The new lung cancer staging system. Chest. 2009;136(1):260-271. [CrossRef] [PubMed]
 
Saji H, Tsuboi M, Shimada Y, et al. A proposal for combination of total number and anatomical location of involved lymph nodes for nodal classification in non-small cell lung cancer. Chest. 2013;143(6):1618-1625. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543