Pulmonary Resection in the Treatment of Life-Threatening Hemoptysis

Abstract

Purpose: Massive hemoptysis is a life threatening situation with high mortality rates. Surgery is effective, however generally an avoided treatment. We report our experience with patients undergoing lung resection for life-threatening hemoptysis.

Methods: Records of all surgically treated patients for hemoptysis between June 2009 and June 2012 were reviewed and analyzed retrospectively.

Results: Anatomical resection was performed on 31 (15.3%) patients out of 203 patients referred to our intensive care unit for life-threatening hemoptysis. 25 (80.6%) were male and six (19.4%) were female; with mean age of 46.4 ± 13.7 (21–77). Pneumonectomy was performed in four (12.9%), lobectomy in 24 (77.4%), segmentectomy in two (6.5%) and bilobectomy in one case. Postoperative complications developed in eight (25.8%), and mortality was observed in two (6.5%) patients. Etiology was bronchiectasis in 13 (42.0%), tuberculosis in eight (25.8%), carcinoma in four (12.9%), aspergilloma in four (12.9%), hydatid cyst in one (3.2%) and lung abscess in one (3.2%) of the cases.

Conclusions: Although lung resection in the treatment of massive hemoptysis is accompanied with high morbidity and mortality rates, surgery is the only permanent curative modality. Acceptable results can be achived in the company of a multidisciplinary approach, through avoidance of pneumonectomy and urgent surgery.

Introduction

Massive hemoptysis is a life-threatening condition. Surgery is effective but surgeons are sometimes reluctant to operate on patients with this condition. Massive hemoptysis is associated with a mortality rate over 50%.^1,2) Several therapeutic strategies have been applied with variable results. Overall mortality, especially after emergent surgery, is still high, ranging from 10% to 38%.¹⁾ A new multidisiplinary approach associating the intensive care physician, the chest physician, the radiologist and the thoracic surgeon enables to lower the mortality rate. In this multidisiplinary approach, the place of surgical treatment among medical treatment and bronchial artery embolization is not obvious.³⁾

In this study we reviewed our experience with patients who underwent pulmonary resections for life-threatening hemoptysis during past three years to determine the indications for surgery and analyze the results of etiology, surgical procedures, the duration between hemoptysis and lung resection, morbidity and mortality rates with possible risk factors.

Material and Methods

Patients

Hospital files of 31 consecutive patients undergoing lung resection out of 203 patients admitted to surgical intensive care unit (ICU) for life-threatening hemoptysis between June 2009 and June 2012 were examined retrospectively. Demographic data of age, gender, quantity of hemoptysis and time to operation, bronchoscopic findings, resection types, pathological diagnosis with postoperative morbidity and mortality ratios were analyzed in terms of comparison. Patients with medically treated and controlled hemoptysis did not undergo any surgical intervention, therefore were excluded from the study.

Definitions

Hemoptysis of more than 200 ml within last 24h, bronchial bleeding causing hemodynamic instability, respiratory impairment, or drop of hematocrit level below 30% was defined as life-threatening hemoptysis. Furthermore, blood losses were subgrouped into “severe” on a daily basis of 200–600 ml and “massive” on an amount over 600 ml/day. Deaths occurred within 30 days after the surgery or before discharge of the patients were defined as operative mortality.

Methods

Fundamentals of the treatment were to prevent the asphyxia, localize and control the bleeding, determine the etiology, and to avoid any recurrence. All patients were managed identically. Conservative treatment immediately initiated in the ICU. After adequate vascular accesses and arterial blood gas tests, oxygen therapy was started. Full blood count, biochemistry and coagulation tests were performed, as well as a cross-match test. Patients are given semi-Fowler’s position with full bed rest. Any arterial hypertension (systolic blood pressure >160 mmHg and/or diastolic blood pressure >90 mmHg) detected was taken under control without delay. Anti-tussives, sedatives and hemostatic drugs, with broad-spectrum antibiotics were initiated routinely. Hemoptysis amounts were monitored cautiously.

All patients underwent rigid bronchoscopy following a chest X-ray. Endobronchial system and the localization of hemorrhage were evaluated. Concomitant interventional methods of cold saline lavage, adrenaline instillation (0.1%), or balloon tamponade were also applied when necessary. In cases with uncontrollable bleeding, selective entubation with a double-lumen endobronchial tube was performed to isolate and to protect the contralateral lung. Hemodynamically unstable patients with resistant bleeding underwent urgent (within the first 72h) or scheduled (>72h) surgical resection subsequent to a detailed radiological examination with computed tomography (CT) scan.

Statistics

Normally distributed, parametrical scale variables were reported as means and standard deviation (SD) with 95% confidence interval (CI) and were compared using Student’s t-test or analysis of variance (ANOVA). Mann-Whitney U test was used for non-parametrical variables. Categorical variables were presented as percentages with ranges and were assessed using Fisher’s exact test or Pearson’s chi-square. In this study, the maximum type I error was 0.05, and the level of significance was accepted as P <0.05. All analyses were performed using SPSS 15.0 for Windows (SPSS Inc., Chicago, Illinois, USA).

Results

There were 25 (80.6%) males and six (19.4%) females, with a mean age of 46.4 ± 13.7. “Mass lesion” was the most common (11/31) pathology detected on preoperative CT scans. Bronchiectasis was evident in seven (22.6%), cavitary lesion in six (19.4%), infiltration in three (9.7%), fibrosis in 1 (3.2%), and pleural effusion in 1 (3.2%) patient. CT scan findings were normal in two (6.4%) patients. Hemoptysis was seen primarily in 17 (54.8%) of the patients.

In terms of hemorrhage quantity, 14 (45.2%) patients were classified as massive and 17 (54.8%) as severe hemoptysis. All patients underwent rigid bronchoscopy, three revealing endobronchial tumor, whereas localization of bleeding was achieved in 26 (83.9%). No pathology was detected in two patients.

Time-to-surgery from the onset of hemoptysis was <24h in 11 (35.5%), 24-48h in four (12.9%), 48-72h in three (9.7%), and >72h in 13 (41.9) patients.

Resection of choice was lobectomy in 24 (77.4%), pneumonectomy in four (12.9%), segmentectomy in two (6.5%), and bilobectomy in one (3.2%) patients. Distributions of operative and postoperative parameters are listed in Table 1 and Table 2.

Table 1.

Distribution of patient and operative demographics

		n	%
Resection type	Lingulectomy	1	3.2
	L lower lobectomy	6	19.4
	L pneumonectomy	2	6.5
	L upper lobectomy	5	16.1
	R lower lobectomy	1	3.2
	R lower lobe superior segmentectomy	1	3.2
	R middle lobectomy	3	9.7
	R pneumonectomy	2	6.5
	R sleeve inferior bilobectomy	1	3.2
	R upper lobectomy	9	29.0
Diagnosis	Tuberculosis	8	25.8
	Bronchiectasis	13	41.9
	Carcinoma	4	12.9
	Aspergilloma	4	12.9
	Other (Hydatid cyst + lung abscess)	2	6.5
Hemoptysis amount	Massive	14	45.2
	Severe	17	54.8
Duration between onset of hemoptysis and operation	0–24h	11	35.5
	24–48h	4	12.9
	48–72h	3	9.7
	>72h	13	41.9
Appearance	Primary	17	54.8
	Recurrent	14	45.2

L: Left; R: Right

Table 2.

Operative procedures for causative diseases

	Bronchiectasis	Tuberculosis	Carcinoma	Aspergilloma	Others	Total
Segmentectomy	1	0	0	1	0	2
Lobectomy	12	6	2	2	2	24
Bilobectomy	0	0	1	0	0	1
Pneumonectomy	0	2	1	1	0	4
Total	13	8	4	4	2	31

Pathological examination of surgical specimens revealed bronchiectasis in 13 (41.9%) patients, tuberculosis (TB) in eight (25.8%), non-small cell lung carcinoma (NSCLC) in four (12.9%), aspergilloma in four (12.9%), and other diagnosis (lung abscess and hydatid cyst) in the further 2 patients. Operative ratios were found to be 67% (4/6) and 42% (13/31) for aspergilloma and bronchiectasis patients, respectively. Proportions of operated patients among the all cohorts are shown in Fig. 1.

Fig. 1.

Distribution of patients’ diagnosis and comparison of operation rates. Footnote → *Others: Arteriovenous malformation, Behçet’s disease, congestive heart failure, excessive anticoagulation, hydatic cyst, hypertention, lung abcess, pneumonia, secondary lung malignancies.

Eight patients (25.8%) developed postoperative complications: prolonged air leak in three, respiratory failure in one, hemorrhage in two, and bronchopleural fistula in two. Overall mortality rate was 6.5% (2/31): One patient undergoing left pneumonectomy for massive hemoptysis due to TB whom was reoperated for postoperative bronchopleural fistula, and another patient undergoing superior sleeve bilobectomy for persistant hemoptysis with NSCLC died on early postoperative course, owing to respiratory insufficiency. A relatively increased morbidity rate was observed in patients undergoing pneumonectomy and lesser resections (50.0% vs 22.2%; P = 0.268). Additionally, no significant correlation was observed between the diagnosis and the postoperative complications (P = 0.741). Distribution and analysis of all complication rates are summarized in Table 3.

Table 3.

Distribution and analysis of complication rates

		Complication
		Yes	No	P
		n (%)	n (%)
Surgical intervention	Others	6 (22.2)	21 (77.8)	0.268
	Pneumonectomy	2 (50.0)	2 (50.0)
Diagnosis	Tuberculosis	2 (25.0)	6 (75.0)	0.741
	Bronchiectasis	3 (23.1)	10 (76.9)
	Carcinoma	2 (50.0)	2 (50.0)
	Aspergilloma	1 (25.0)	3 (75.0)
	Others	0 (0)	2 (100.0)
Hemoptysis amount	Massive	6 (42.9)	8 (57.1)	0.097
	Severe	2 (11.8)	15 (88.2)
Duration between onset of hemoptysis and operation	0–24h	3 (27.3)	8 (72.7)	0.985
	24–48h	1 (25.0)	3 (75.0)
	48–72h	1 (33.3)	2 (66.7)
	>72h	3 (23.1)	10 (76.9)
	<72h	5 (27.8)	13 (72.2)	1.000
	>72h	3 (23.1)	10 (76.9)
Appearance	Primary	4 (23.5)	13 (76.5)	1.000
	Recurrent	4 (28.6)	10 (71.4)

Complication rates of patients with massive and severe hemoptysis were 42.9% and 11.8%, respectively (P = 0.097). In addition, no significant correlation was observed between the complication rates and the time-to-operation duration. Patients operated within the first 24 h, between 24–48 hours, 48–72 h, and after 72 h developed complications in 27.3%, 25.0%, 33.2%, and 23.1%, respectively (P = 0.985). Postoperative complication rate did not differ between patients with recurrent or primary hemoptysis attack (P = 1.000).

Distribution of mortalities among different parameters is listed in Table 4. Since there were only 2 cases, no statistical analysis was possible for correlation. Operative rate and 90-day hemoptysis-related mortality for each cohort is given in Table 5.

Table 4.

Distribution of mortalities

		Mortality
		Yes	No
		n (%)	n (%)
Surgical intervention	Pneumonectomy	1 (25.0)	3 (75.0)
	Others	1 (3.7)	26 (96.3)
Diagnosis	Tuberculosis	1 (12.5)	7 (87.5)
	Bronchiectasis	0 (0)	13 (100.0)
	Carcinoma	1 (25.0)	3 (75.0)
	Aspergilloma	0 (0)	4 (100.0)
	Others	0 (0)	2 (100.0)
Hemoptysis amount	Massive	1 (7.1)	13 (92.9)
	Severe	1 (5.9)	16 (94.1)
Duration between onset of hemoptysis and operation	0–24h	0 (0)	11 (100.0)
	24–48h	1 (25.0)	3 (75.0)
	48–72h	0 (0)	3 (100.0)
	>72h	1 (7.7)	12 (92.3)
	<72h	1 (5.6)	17 (94.4)
	>72h	1 (7.7)	12 (92.3)
Appearance	Primary	1 (5.9)	16 (94.1)
	Recurrent	1 (7.1)	13 (92.9)

Table 5.

Distribution of operative ratios and 90-day hemoptysis-related mortality rates for different cohorts

Definitive diagnosis	Total number of patients	Operated patients		Non-operated patients
		n (%)	Mortality (%)	n (%)	Mortality (%)
Primary lung malignancy	40	4 (10)	1 (25)	36 (90)	19 (53)
Tuberculosis	42	8 (19)	1 (13)	34 (81)	3 (9)
Bronchiectasis	31	13 (42)	0 (0)	18 (58)	2 (11)
Aspergilloma	6	4 (67)	0 (0)	2 (33)	0 (0)
Others*	39	2 (5)	0 (0)	37 (95)	9 (24)
No definitive diagnosis	45	0 (0)	0 (0)	45 (100)	4 (9)

^*Others: Arteriovenous malformation, Behçet’s disease, congestive heart failure, excessive anticoagulation,hydatic cyst, hypertention, lung abcess, pneumonia, secondary lung malignancies

Two patients diagnosed as NSCLC died of non-hemorrhagic causes on 8th and 12th months, postoperatively. Recurrence of hemoptysis was observed in only one patient (3.2%). This patient with diagnosis of bronchiectasis underwent right lower lobectomy 4 months after his first middle lobectomy operation. The follow up period ranged from 1 to 4 years. Median follow-up period of our cohort was 23 months.

Discussion

Definition of massive hemoptysis has a wide range of difference in terms of hemorrhage volume (200–1000 ml/day) in the literature.^1,2,4) As an alternative, hemoptysis should be classified due to its severeness and life-threatening circumstance instead of its quantity. In our series of patients undergoing lung resection for life-threatening hemoptysis, 17 (54.8%) had a bleeding amount of 200–600 ml/day, and the rest 14 (45.2%) had an amount over 600 ml/day. In view of the fact that there was over three-fold more postoperative complications seen in the massive hemoptysis group according to the severe hemoptysis group, this difference did not reach the significance level (42.9% vs. 11.8%; P = 0.097).

Hemorrhage into the tracheobronchial system may be originated from bronchial or pulmonary artery networks. Bleedings from bronchial artery system usually occur as a consequence of neovascularization, and accompany inflammatory pulmonary diseases such as bronchiectasis, micobacteriosis, or other suppurative lung diseases. These hypertrophic neovascularizations have musculature wall allowing vasoconstriction. Interventions causing vasospasm such as percutaneous embolization or other pharmacological methods may provide a temporary reduce or cessation of the hemorrhage, but surely will not be permanent until the underlying disease persists.^2,3,5)

Pulmonary artery system does not have a muscular wall producing vasospasm, as in bronchial arterial system. Instead they have thin walls and may cause massive and usually fatal hemorrhage due to the mechanical damage formed by lung cancer, aspergilloma, or necrotizing pneumonia. In such cases, no time should be wasted by non-surgical interventions. As Jougon et al. have also stated,³⁾ however, that it is not always possible to differentiate the hemoptysis in these two categories, and predict the prognosis. For this reason, surgical resection has been the most vital treatment of choice.

In a study by Knott-Craig, et al., 36.4% of all patients admitting to the hospital for massive hemoptysis and non-surgically treated are found to re-admit within the following 6 months, and 45% of these recurrences were fatal.⁶⁾ When compared with conservative treatment modalities, surgery allows curative treatment for primary hemoptysis and recurrences. In our series, only one patient (3.2%) had recurrence after surgical resection.

Bronchial artery embolization has been gaining popularity during the last decade. In two different studies by Haponik, et al. in 1990 and 2000, the preference of clinicians on interventional angiography in life-threatening hemoptysis has raised from 21% to 50%.^7,8) Microcatheters allowing superselective catheterization, recently discovered embolic materials, increased experience in the interventional radiology, and low rate of reported complications are thought to play the major role in this increased recognition.

Treatment of massive hemoptysis should take place in ICU, under cautious monitorization. There is no consensus, yet, in the current literature on the timing of bronchoscopy or the surgical resection. But there are some recommendations on performing surgery subsequent to embolization.^3,9) Since the hemorrhage comes from the bronchial artery system in majority of patients, embolization may control the hemoptysis in 75%-90% of patients, however through a recurrence rate up to 75%.^3,5,10) From this perspective, bronchial artery embolization can facilitate a more detailed assessment of the patient and a safer operative course. On the reason of being a pulmonology hospital without an interventional radiology unit, we were unable to perform this method on our patients, and could not refer any patient to other hospitals due to their relatively instable vital conditions.

Morbidity and mortality rates are found to be higher in emergency cases than elective pulmonary resections for hemoptysis.^1,4,11) In patients necessitating urgent resections, the pulmonary reserve is usually compromised due to aspiration, and the resection which can be extensive, is performed during the bleeding. Andrejak, et al. have reported a series of 111 (13.6%) lung resections for severe hemoptysis out of 813 cases admitted to ICU,¹¹⁾ in accordance to our surgery rate of 15.8% (31/208). They have categorized resections into three groups as emergency, scheduled (after bleeding control) and planned (after discharge). Mortality ratios were 35%, 4% and 0% among the groups, respectively. This study indicated the importance of avoiding emergency resections as much as possible, utilizing non-surgical methods to stop the hemorrhage and optimize the circumstances for the patients prior to any resection, in order to minimize the morbidity and mortality. Likewise, Shigemura, et al. have stated that patients undergoing emergency resections for life-threatening hemoptysis have high morbidity and mortality rates according to patients undergoing preoperative bronchial artery embolization.¹²⁾ Overall morbidity rate of our cohort was 25.8%. Despite the high postoperative complication rate seen in pneumonectomy patients, no significant difference was observed between complication rates of pneumonectomy and lesser resections (50.0% vs. 22.2%; P = 0.268), probably due to the unjustification of the low number of our cases.

Optimistically, Metin, et al. have reported reasonable results of 27.5% morbidity and 11.5% mortality in 29 patients (28 anatomical resections, one exclusion) undergoing emergency surgery for massive hemoptysis,⁴⁾ similar to our rates of 25.8% morbidity and 6.5% mortality. The difference between mortality rates may show a discrepancy according to performing all operations under urgent circumstances. In our series, 41.9% of patients were operated after more than 72 h of the onset of hemoptysis; and still no significant difference was observed between mortality rates of delayed and emergency operations, probably as a cause of our low patient volume.

Hemoptysis may be seen during the course of various diseases; bronchiectasis, TB, pulmonary malignencies, pneumonia, and lung abcess being the most frequent causes. The frequencies of these diseases vary according to geographical location, socioeconomical status of the patients and the time period of which the study was conducted. Etiological factors and their frequencies reflect this broad dispersion.^13–15)

Bronchiectasis (41.9%), TB (25.8%), and aspergilloma (12.9%) were the major histopathological diagnosis after lung resection for massive hemoptysis in our study. No significant difference of morbidity or mortality rate was seen in between patients with different pathologies (P = 0.741). In a cohort study by Hirsberg, et al., most common etiologies for hemoptiysis were bronchiectasis (20%) and pulmonary malignancy (19%), where TB was reported to be a rare (1.4%) factor.¹⁴⁾ Despite its sporadic appearance in Western countries, TB still holds its infamous reputation among other etiological factors of life-threatening hemoptysis in developing countries. In a study from our country, TB has been reported the most frequent (34.5%) cause of massive hemoptysis in patients undergoing emergency lung resections.⁴⁾ Correspondingly, Erdogan et al. have also affirmed the high prevalence of TB in hemoptysis patients, especially in countries where the disease is endemic. With a mortality of 6.8% and no recurrence seen in the cohort, it has been concluded that surgery is a safe and permanent curative treatment for massive hemoptysis cases.¹⁶⁾ Correspondingly, our results revealed relatively low mortality among operated patients compared with non-operated patients, indicating the importance and curative intent of surgery in accurately selected hemoptysis patients (Table 5). Terminal stage patients with primary lung carcinoma, medically or surgically unsuitable for a resection was thought as the major cause of high mortality seen in non-operated patient group. Pulmonary resection has also been recommended to treat and prevent further complications of TB leading to hemoptysis, such as middle lobe syndrome, broncholithiasis, or post-obstructive bronchiectasis.¹⁷⁾

Conclusion

Despite the surgical resection is the only curative treatment of choice for massive hemoptysis patients, it should be used as a fraction of a multidisciplinary approach. Emergency or elective anatomical lung resections can be performed safely with reasonable morbidity and mortality rates; however the urgent interventions may carry an increased risk as a consequence of imprecise evaluation of pulmonary functions and other comorbidities. Total surgical resection must be the treatment of choice for all patients with localized causative lesions, especially like bronchiectasis, only if the removal of all involved lung parenchyma is radiologically feasible. Palliative resection may be considered also for patients with life-threatening hemoptysis and inoperable or terminal-stage lung carcinoma, if the hemorrhage localization can completely be excised. Available interventional therapies such as bronchial blockers or arterial embolization should be used in an optimal manner preoperatively, in order to optimize the conditions for the operation.

Disclosure Statement

None.