Hodgkin's disease

Study objectives: To study the incidence of upper airway obstruction, as measured on the flow volume loop (FVL), in patients with bulky mediastinal Hodgkin's disease; to correlate the FVL with CT of the chest; and to follow the changes in the FVL after treatment of the tumor.

Design: Retrospective study of pulmonary function tests (PFTs) and chest CTs performed as part of a clinical trial for Hodgkin's disease.

Setting: Memorial Sloan-Kettering Cancer Center, a comprehensive cancer care center.

Patients: Twenty-five patients (15 men and 10 women; age range, 20 to 57 years) with bulky mediastinal Hodgkin's disease enrolled in a clinical trial of chemotherapy followed by external beam radiation therapy.

Measurements and results: Fourteen of 25 patients (56%) had an abnormal FVL prior to therapy; after chemotherapy, only 7 of 25 patients (28%) had an abnormal FVL. The abnormal patterns seen were either those typical of fixed obstruction or variable extrathoracic obstruction. No patient had a pattern typical of variable intrathoracic obstruction. On chest CT scan, 16 patients had grade-I tracheal deformity; 6 had grade-II deformity, and 3 had grade-III deformity. All three patients with grade-III deformity had a fixed obstruction pattern, as did three patients with a grade-I pattern. Patients with a fixed pattern on FVL had significant decreases in inspiratory and expiratory flow rates.

Conclusion: FVL abnormalities suggesting upper airway obstruction occurred in > 50% of patients with bulky mediastinal Hodgkin's disease. A fixed pattern of obstruction was associated with the lower flow rates and severe tracheal distortion on CT; these patients may warrant special attention prior to general anesthesia or invasive procedures. Asymptomatic patients with abnormal FVLs but normal tracheal profiles need not undergo extensive evaluation. No patients showed the expected pattern typical of intrathoracic obstruction, but rather the major effect was on the inspiratory loop. The authors speculate on the mechanism for this unexpected finding.

(CHEST 2000; 117:1256-1261)

Key words: Hodgkin's disease; pulmonary function testing; upper airway obstruction

Abbreviations: ET = extrathoracic; [FEF.sub.50]/[FIF.sub.50] = forced expiratory flow rate at 50% of vital capacity/forced inspiratory flow rate at 50% of vital capacity; FVL = flow volume loop; HD = Hodgkin's disease; IT = intrathoracic; PFTs = pulmonary function tests; UAO = upper airway obstruction

Hodgkin's disease (HD) typically presents with lymphadenopathy. The mediastinum is most commonly involved, and adenopathy is found in this location in up to 60% of patients at presentation.[1] Hodgkin's-related adenopathy can be massive, and there are reports of life-threatening upper airway obstruction (UAO) from this cause.[2] Compression of the airway has been described most commonly when a patient with bulky adenopathy is receiving general anesthesia or undergoing an invasive procedure. The incidence of severe upper airway compromise at presentation in patients with HD has been estimated to be as high as 2.4%,[2] but the exact frequency in this setting has not been systematically studied.

CT scans of the chest allow a detailed anatomic view of the airways. Pulmonary function tests (PFTs) permit dynamic assessment of airway obstruction throughout the respiratory cycle. The flow volume loop (FVL) is particularly helpful in identifying the presence of UAO. These two studies are regularly done to assess disease and effects of therapy on the patients with HD at our center. We became aware of frequent abnormalities suggesting UAO on PFTs in patients with bulky mediastinal disease receiving chemotherapy. We therefore undertook this study to characterize the incidence, type, and severity of UAO in patients with bulky HD in the mediastinum, and the influence of chemotherapy on these findings. We analyzed the physiologic parameters of airway obstruction using PFTs and correlated them with anatomic findings on CT.

MATERIALS AND METHODS

Patients

Patients included were those with early bulky HD who were prospectively enrolled in a research protocol at Memorial Sloan-Kettering Cancer Center from June 1991 to August, 1994. The purpose of that study was to determine the effects of involved vs extended field radiation therapy administered after chemotherapy for patients with early stage HD with bulky lymphadenopathy. Bulky disease in the chest is defined as a mediastinal mass greater than one third of the thoracic diameter; patients had either IA/B, IIA/B, or IIIA[3] disease. All patients received doxorubicin, bleomycin, vinblastine, and dacarbazine followed by either involved field or extended field radiation. Involved field radiation therapy was treatment to the bulky disease and other involved lymph node sites; extended field therapy was radiation to the mantle portal, inverted Y portal, or combinations of both.[4] This study had been approved by the Institutional Review Board of Memorial Sloan-Kettering Cancer Center, and all patients gave informed consent. Thirty-six patients with bulky HD were enlisted on this protocol. Twenty-five of these patients had both baseline and follow-up PFTs for review. The mean age of the patients was 31 years (range, 20 to 57 years). There were a total of 15 men and 10 women. HD was of the nodular sclerosis type in 23 patients and of the mixed cellularity type in 2 patients.

PFTs

PFTs were obtained at baseline and at 4 months after chemotherapy. All subjects had values for spirometry, peak inspiratory flow rate; maximal forced expiratory flow rate, as percent of predicted; forced expiratory flow rate at 50% of vital capacity/ forced inspiratory flow rate at 50% of vital capacity ([FEF.sub.50]/ [FIF.sub.50]); and [FEV.sub.1] as percent of predicted recorded. PFTs were performed on the Collins Gold standard (GS/Plus; Warren E. Collins; Braintree, MA) system using standard American Thoracic Society procedures. All patients had an FVL test performed. The FVLs were interpreted using criteria of Miller and Hyatt.[5] Patients with a plateau only on inspiratory flow were considered to have a pattern consistent with variable extrathoracic (ET) UAO, while those with both flattening of the inspiratory and expiratory flows were considered to have a pattern suggestive of a fixed obstruction. A plateau of the expiratory flow with a normal inspiratory flow was considered to be a pattern suggestive of an intrathoracic (IT) UAO. The FVLs were interpreted independently by three pulmonologists (DAW, NVE, and FS) without knowledge of the CT scan results. Results were discussed, and a consensus interpretation was agreed on.

Radiologic Evaluation

The CT scan of the chest that corresponded temporally to the PFTs was reviewed by a radiologist (AMB) without knowledge of the PFTs. The tracheal distortion was graded as I, II, or III. Grade I was defined as an IT trachea showing no or minimal distortion and being round and symmetrical in shape. Grade II demonstrated moderate distortion and was oval in shape. Grade III was defined as a trachea demonstrating severe distortion and narrowing, with a large discrepancy ([is greater than] 10 mm) between transverse and anteroposterior dimensions.

Statistics

All results, unless otherwise stated, are expressed as mean [+ or -] SEM. The Student's t test was done to test for statistical significance in the spirometric measurements that were performed in those with UAO compared to normal subjects. Analysis of variance was done to evaluate statistical significances between subgroups. A value of p [is greater than] 0.01 was taken as the level of statistical significance.

RESULTS

Analysis of the initial FVL of the 9,5 study patients showed that at presentation prior to treatment, 14 were abnormal (56%), while 11 were normal (44%; Table 1). The results are listed for individual patients in Table 2. Two patterns of abnormality were seen. Seven patients had flattening of both their inspiratory and expiratory loops, giving a picture suggestive of fixed obstruction. Another seven patients (28%) had flattening only on the inspiratory loop, a pattern usually associated with an ET obstruction. No patient had flattening only on the expiratory loop in spite of bulky IT mediastinal disease in all cases. After chemotherapy, normal FVL patterns were seen in 19 patients (76%). Of the six patients (24%) with persistent abnormalities, five had flattening on the inspiratory loop only, and one had a fixed pattern (Table 1). The flow volume curve before and after chemotherapy in patient 16 is shown in Figure 1. FVLs from patients 9, 11, 15, and 23 are shown in Figure 2.

[Figures 1-2 ILLUSTRATION OMITTED]

(*) ET = extrcthoracic;

IT = intrathorcric.

Table 2--Trachea Grade and FVL Patterns for Each Patient

Spirometric data for all patients are given in Table 3. The [FEV.sub.1] percent predicted was unaffected and was similar in all groups, and showed no change with chemotherapy. The actual flow rates as a percentage of predicted were significantly decreased on both inspiration and expiration in those with a fixed pattern, compared to other groups. Those with an inspiratory pattern had lower absolute mean inspiratory flows, but this did not reach statistical significance. In all cases, the [FEF.sub.50]/[FIF.sub.50] was high, suggesting that even when both limbs were affected, the impact was greatest on the inspiratory loop.

Table 3--Spirometric Values in Patients With Bulky Mediastinal Disease(*)

(*) Data are presented as mean [+ or -] SEM; % pred = percent predicted; FEFmax = maximal forced expiratory flow rate; PIFR = peak inspiratory flow rate.

([dagger]) p < 0.001 compared to normal subjects.

([double dagger]) p < 0.001 compared to before chemotherapy.

Analysis of CT scans showed that 16 patients (64%) had grade-I changes, 6 patients (24%) had grade-Il changes, and 3 patients (12%) had grade-III changes. The relationship of anatomic changes seen on scans to FVL abnormalities is shown in Figure 3. All three patients with very severe tracheal distortion had fixed patterns on the FVL. In two of these cases, the trachea was displaced to the right; in the remaining case, the trachea was normal in position. The measurements at the site of maximum narrowing of the IT trachea in anteroposterior and transverse are as follows: 25 x 8 mm, 28 x 12 mm, and 20 x 8 mm. The areas of tracheal compression were IT in all cases.

[Figure 3 ILLUSTRATION OMITTED]

No patient experienced clinically significant respiratory distress due to UAO during the period of the study.

DISCUSSION

HD, particularly of the nodular sclerosis subtype, frequently presents with bulky mediastinal disease. Although life-threatening UAO is rare, our study demonstrates that tracheal distortion on CT and physiologic impairment demonstrated by PFTs are common findings. Moderate or severe tracheal abnormality was present on 36% of scans, and physiologic abnormalities were detected in PFTs prior to treatment in 56%.

As UAO worsens, peak expiratory or inspiratory flows are affected early, while the [FEV.sub.1] is affected only at a latter stage.[6,7] Thus, spirometry alone is often not helpful in detecting the presence of UAO. This was the case with our patients all having a normal [FEV.sub.1] percent predicted. Distortion of the flow volume curve is more helpful in suggesting UAO.

In patients with the most severe (grade III narrowing of airways noted on CT, a fixed pattern of obstruction was noted, although it was also seen in three patients with a grade-I pattern. This pattern was with a decrease in absolute flow rates throughout the entire maneuver. Notably in all cases, the [FEF.sub.50]/ [FIF.sub.50] was high, suggesting a more predominant effect on inspiration. This was not expected with IT disease. Some of our patients also had lymphadenopathy in the neck that may have contributed to these patterns, but none had evidence of ET tracheal obstruction on CT, and the predominant site of lymphadenopathy disease was in the chest. Also, the patients with abnormal loops who did not have a fixed pattern were also found to have their greatest air flow limitations during inspiration rather than expiration. Even when both limbs were affected, the severity was greatest on the inspiratory limb.

This unique finding has not been previously reported with bulky adenopathy. We postulate that this is because our patients did not have true intra-airway obstruction, but rather external splinting of the airways. We speculate that the adenopathy prevented the airways of some patients from fully expanding in response to the negative IT pressure during inspiration, thus leading to lower flow rates on the inspiratory curve. During expiration, these nodal masses, which were rigid and fixed in place, generally did not further compress the airways; in response to positive pressure generated during expiration. An alternative hypothesis would be that the trachea, made more rigid by the adenopathy, more effectively transmitted negative pressures to the airway above the manubrium, exaggerating the effect of inspiration on the proximal trachea, larynx, and pharynx.

We observed no episode of clinically significant respiratory distress, so it is impossible to predict the likelihood of a respiratory problem during anesthesia or invasive procedures from this study. We believe, however, that the patients at greatest risk for these complications are likely those with a fixed obstruction pattern on FVLs, because of the significantly decreased flow rates and distortion on chest CT. The clearest utility of our study may be to reassure clinicians who encounter abnormal FVLs in asymptomatic patients and normal trachea profiles; an extensive workup may be avoided.

In future studies, measurement of airway impedance would be helpful to characterize the pathophysiology of airflow obstruction. The longitudinal extent of obstruction should also be characterized, perhaps by evaluation of multiple transverse cuts, in addition to evaluation of tracheal cross-section.

In summary, we have found there are frequent abnormalities on the FVL of patients with bulky mediastinal HD, which are correlated with the degree of tracheal distortion. It improves in most cases with chemotherapy. The presence of a pattern of both flattened inspiratory and expiratory loops suggests severe tracheal narrowing and a potentially high-risk patient. We also describe the previously unseen effect of bulky mediastinal adenopathy causing flattened inspiratory flow loops.

REFERENCES

[1] Selby P, McElwain TJ. Clinical features of Hodgkin's disease. Oxford, UK: Blackwell, 1987; 94-125

[2] Jeffery GM, Mead GM, Whitehouse JM. Life-threatening airway obstruction at the presentation of Hodgkin's disease. Cancer 1991; 67:506-510

[3] Carbone PP, Kaplan HS, Mushhoff K, et al. Report of the committee on Hodgkin's disease classification. Cancer Res 1971; 31:1860-1861

[4] Kaplan HS. The radical radiotherapy of regionally localized Hodgkin's disease. Radiology 1962; 78:553-561

[5] Miller RD, Hyatt RE. Evaluation of obstructing lesions of the trachea and larynx by FVL. Am Rev Respir Dis 1973; 108:475-481

[6] Miller RD, Hyatt RE. Obstructing lesions of the larynx and trachea: clinical and physiologic characteristics. Mayo Clin Proc 1969; 44:145-161

[7] Kryger M, Bode F, Antic R, et al. Diagnosis of obstruction of the upper and central airways. Am J Med 1976; 61:85-93

(*) From the Pulmonary Service (Drs. White, Vander Els, and Sorhage) and the Lymphoma Service (Dr. Straus) of the Department of Medicine, and the Department of Radiology (Dr. Bach), Memorial Sloan-Kettering Cancer Center and Cornell University Medical College.

Manuscript received March 5, 1999; revision accepted November 10, 1999.

Correspondence to: Nicholas J. Vander Els, MD, FCCP, 1275 York Ave, New York, NY 10021; e-mail: vanderen@mskcc.org

COPYRIGHT 2000 American College of Chest Physicians
COPYRIGHT 2000 Gale Group