Abstract
Background:
Pleural effusions are a common complication of end-stage renal disease. These effusions are occasionally refractory to medical management, but few options are then available. Indwelling pleural catheter insertion (IPC) has been well described for the management of malignant pleural effusions and, more recently, of nonmalignant effusions of other origin. We aimed to analyze our experience and to evaluate the safety and feasibility of using IPCs for pleural effusion associated with end-stage renal disease.
Methods:
We constructed a cohort of patients who underwent IPC insertion for pleural effusions associated with end-stage renal disease. The IPCs were inserted as a palliative measure in patients who had thoracentesis twice within the preceding 2 weeks, no evidence of infection and either failure to respond, complications or intolerance to maximal medical therapy, or if IPC insertion would enable discharge when the patient was hospitalized mainly for dyspnea due to pleural effusion.
Results:
There were nine IPCs inserted in eight patients. Patients had significant dyspnea at baseline with a median baseline dyspnea index of 1.5 [interquartile range (IQR) 0–3]. Dyspnea improved significantly 2 weeks after catheter insertion with a median transitional dyspnea index of 6 (IQR 4.5–7.0). There was no occurrence of empyema or other major complications. Serum albumin did not decrease after catheter insertion. IPCs were removed in four patients (50%) and successful spontaneous pleurodesis occurred in three patients (37.5%) after a median of 77 days (IQR 9–208).
Conclusion:
IPC insertion for pleural effusions associated with end-stage renal disease appears safe and effective. Larger studies are needed, particularly regarding the impact of this intervention on quality of life.
Keywords
Introduction
Hemodialysis has transformed the life expectancy of those who have end-stage renal disease (ESRD). That transformation brings a surge of complications associated with chronic renal failure, one complication being pleural effusions. Pleural effusions are common in patients with ESRD. One retrospective analysis of 257 patients who received long-term hemodialysis between 1990 and 2006 found the incidence of pleural effusions in this population to be as high as 20%. The etiology of these effusions is multifactorial and included hypervolemia (61.5%), heart failure (9.6%), parapneumonic effusion (9.6%) and uremic pleuritis (3.8%) [Bakirci et al. 2007]. However, this analysis did not include patients on peritoneal dialysis or those who did not yet need dialysis.
Current strategies for the management of refractory pleural effusions in ESRD are limited. Serial therapeutic thoracenteses are sometimes performed but expose the patient repeatedly to procedural risks. Pleurodesis is also sometimes performed, but there is little evidence for its use for nonmalignant pleural effusions in the literature, and even less so when the pleural effusion is due to ESRD [Jenkins and Shelp, 1974]. Furthermore, in the few reported cases of pleurodesis for cardiogenic pleural effusions, concerns were raised about the safety of this procedure [Spicer and Fischer, 1969; Davidoff et al. 1983].
Placement of an indwelling pleural catheter (IPC) with intermittent outpatient drainage by the patient or a patient attendant is an accepted treatment for patients with recurrent malignant effusions [Tremblay and Michaud, 2006; Warren et al. 2008]. It provides the advantage of shortening length of stay as the procedure can be done in the outpatient setting and can also be used in cases where there is irremediable lung entrapment [Pien et al. 2001].
We have recently shown that the use of IPC for the management of cardiogenic pleural effusion was feasible and associated with few complications [Srour et al. 2013b]. Various other nonmalignant pleural effusions such as those due to hepatic hydrothorax, inflammatory pleuritis and chylothorax have been successfully treated with an IPC with a low complication rate and high patient satisfaction [Chalhoub et al. 2011; Bhatnagar et al. 2014] However, there are no data regarding the use of IPCs for pleural effusions secondary to ESRD. We therefore aimed to present our own experience with this management option and determine whether IPC insertion for pleural effusions secondary to ESRD is feasible and safe.
Methods
Study design and setting
We identified a cohort of patients who underwent IPC insertion (PleurX®, CareFusion, San Diego, CA, USA) for a pleural effusion due to renal disease at the Chronic Ascites and Recurrent Effusion (CARE) clinic at The Ottawa Hospital using the clinic’s prospectively constructed registry.
Ottawa’s CARE Clinic, established in May 2006, was originally started as a malignant pleural effusion clinic. The clinic is run by an interprofessional group including an interventional pulmonologist, an advanced practice nurse specialized in palliative care and a clinic nurse specialized in malignant effusions. The clinic now addresses refractory nonmalignant pleural effusions. The investigational protocol was reviewed and approved by the Ottawa Health Science Network Research Ethics Board as protocol number 2011873-01H. The Board did not require informed consent to be obtained.
Participants
Patients who underwent IPC insertion for a pleural effusion due to renal failure between May 2006 and January 2012 were included. Patients who underwent IPC insertion during pleuroscopy for talc pleurodesis were excluded from statistical analysis but are reported on. Patients were required to have failed other treatments to undergo IPC insertion. Specific criteria were as follows: A (1) failure to respond to maximal medical therapy (Lasix 160mg/day, spironolactone 400 mg/day), or (2) complications or intolerance to medical therapy such as worsening renal failure or hypotension during dialysis limiting fluid removal, or (3) hospitalization mainly for management of dyspnea due to the effusion such that insertion of an IPC would facilitate discharge; and (B) thoracentesis twice within the preceding 2 weeks; and (C) no evidence of an infection such as spontaneous bacterial peritonitis or pneumonia.
The IPC catheter was inserted as a palliative measure. The IPC was removed when the drainage was less than 50 ml on three consecutive visits, unless the patient reported increased dyspnea and there was radiological evidence of pleural fluid re-accumulation on imaging suggestive of catheter blockage or loculation.
Data sources
Most data were collected prospectively in the CARE clinic database. Additional data were retrospectively obtained from medical records including renal failure etiology, pleural effusion pathophysiology, Eastern Cooperative Oncology Group (ECOG) performance status, survival after the index procedure, and albumin levels. Additionally, chest X-rays were visualized through the Ottawa Hospital’s picture archiving and communication system.
Outcomes
The primary outcome was the improvement in dyspnea evaluated by the baseline dyspnea index (BDI) and transitional dyspnea index (TDI) 2 weeks after IPC insertion. The BDI has three components graded from 0 (very severe) to 4 (no impairment) and therefore ranges from 0 to 12. The TDI has three components graded from –3 (major deterioration) to +3 (major improvement) and therefore ranges from –9 to +9. A TDI of 1 unit is the minimum clinically important difference [Mahler and Witek, 2005]. Other outcomes included lung re-expansion, change in serum albumin, catheter removal and fluid recurrence, depletion of protein stores and the need for a second pleural intervention.
The index procedure was identified as the initial IPC placement. Adequate lung re-expansion after a pleural procedure was defined as at least 80% pleural apposition on the affected hemithorax as determined visually on the postprocedure chest radiographs and partial re-expansion was defined as pleural apposition between 50 and 80%. Recurrence of an effusion was defined as recurrence of an effusion occupying at least 50% of the hemithorax.
Complications were also assessed including unsuccessful insertion, symptomatic loculation, asymptomatic loculation, empyema, pneumothorax, subcutaneous emphysema, bronchopleural fistula, cellulitis, blocked catheter, dislodged catheter, bleeding, tumor seeding, pain requiring removal, acute respiratory distress syndrome (ARDS), transient respiratory distress, fever or leak.
Management of IPC
All IPCs were inserted under local anesthesia with ultrasound guidance. After IPC insertion all patients were provided with subsequent nursing services at hemodialysis for pleural drainages or via homecare on a schedule individualized to each patient. Nurses received special training in the management of the catheter and were provided with a troubleshooting algorithm. Both patients and staff were encouraged to contact a dedicated pleural effusion clinic nurse with questions or concerns regarding the catheter. Patients were seen in follow up 2 weeks following catheter insertion and then every 6 weeks or as required until catheter removal or patient death. During these appointments, symptom control was assessed using BDI and TDI.
Statistical analysis
We used SAS 9.3 statistical software (SAS Institute Inc., Cary, NC, USA). All analyses were conducted with patients as the unit of analysis, considering only the first IPC insertion. Quantitative data were reported as median and interquartile range (IQR). The change in albumin level after IPC insertion was analyzed using the Wilcoxon rank sum test. Survival analysis was performed using the Kaplan–Meier method.
Results
There were nine IPCs inserted in eight patients (Table 1) including one who had a subsequent ipisilateral IPC insertion. There were three additional patients on peritoneal dialysis who had an IPC inserted during pleuroscopy for talc pleurodesis due to patient preference. The catheter was removed after 14 days. These three patients were not included in the analysis because the talc pleurodesis made a large contribution to the pleurodesis or improvement of symptoms.
Baseline characteristics at the time of first IPC insertion.
Data are presented as n (%) or median (IQR) as appropriate.
BDI, baseline dyspnea index; ECOG, Eastern Cooperative Oncology Group; GFR, glomerular filtration rate; IPC, indwelling pleural catheter; IQR, interquartile range.
At the time of first IPC insertion, the median age of the eight patients was 82.2 years (IQR 72.1–87.7). There were five males (62.5%) and three females (37.5%). Performance status was poor, with four patients (50%) having an ECOG performance status of 4 and three patients (37.5%) having an ECOG performance status of 3. There were as many left-sided as right-sided effusions, and as many transudative as exudative effusions according to Light’s criteria [Romero et al. 1993]. The IPC was inserted in two patients (25.0%) during pleuroscopy. The median amount of fluid drained immediately after IPC insertion was 1.60 l (IQR 0.65–2.20). Drainage was stopped when there was no more fluid draining, but in one case drainage was limited by pain.
Etiologies of the ESRD included diabetic nephropathy (37.5% of patients), glomerulonephritis not otherwise specified (37.5%), interstitial nephritis (12.5%) and nephrotic syndrome (12.5%). The pleural effusion pathophysiology was diastolic heart failure and volume overload in three patients (37.5%), uremia in two patients (25%), pleural–peritoneal communication while on peritoneal dialysis in one patient (12.5%) and undetermined in two patients (25%). One patient was on peritoneal dialysis (12.5%) and six (75%) on hemodialysis. One patient had a glomerular filtration rate (GFR) of 15 ml/min/1.73 m2 but was not yet on dialysis. The patient on peritoneal dialysis declined both switching to hemodialysis and pleurodesis.
Patients had significant dyspnea prior to the procedure, with four patients having a BDI of 0 and 4 patients with a BDI of 3 (median 1.5, IQR 0–3). Dyspnea improved significantly 2 weeks after catheter insertion with a median TDI of 6 (IQR 4.5–7.0). All patients achieved a TDI of 3 or more whereas a TDI of 1 is the minimal clinically important difference. There were no instances of worsening dyspnea. There was adequate lung re-expansion on the chest radiograph in five cases (62.5%) and partial re-expansion in three cases (37.5%). Albumin, measured a median of 19 days before catheter insertion, was decreased at baseline at a median of 26.5 g/l (IQR 25.0–30.5). Albumin level, measured again at a median of 34 days after catheter insertion, was found to be at a median of 28.0 g/l (IQR 25.5–30.0). The median change in albumin level was 1.0 (IQR -1.5 to 2.0; p = 0.67).
IPCs were eventually removed in four patients (50%, Table 2) after a median of 45.5 days (IQR 11.5–142.5). This included two patients with transudative effusions (50% of patients with transudative effusions) and two patients with exudative effusions (50% of patients with exudative effusions). There was subsequent recurrence of an effusion at least moderate in size in one patient from the transudative effusion group, 3 days after the catheter was removed due to being dislodged. Thus successful pleurodesis defined as catheter removal without subsequent recurrence of a pleural effusion at least moderate in size occurred in three patients (37.5%) after a median of 77 days (IQR 9–208). There were no instances where further pleural interventions such as pigtail or large-bore tube thoracostomy, pleurodesis, pleuroscopy or video-assisted thoracic surgery were needed. One catheter needed to be reinserted after being dislodged. Other complications were minor and included subcutaneous emphysema and re-expansion pulmonary edema (one patient each). There were no occurrences of empyema or any other major complications.
Re-expansion and catheter removal after IPC insertion.
Data are presented as n (%).
IPC, indwelling pleural catheter.
Following IPC insertion, the admission rate was 1.22 per year [95% confidence interval (CI) 0.68–2.20]. Median follow up was 184.5 days (IQR 72–704.5). Of the eight patients, four expired. Median survival was 318 days (Figure 1, IQR 28 to ‘not reached’).
Survival.
Discussion
Our data shows the use of IPCs for refractory effusions associated with ESRD is effective in relieving dyspnea in selected patients. Lung re-expansion with pleural apposition >80% was achieved in the majority of patients and the catheter could be removed without recurrence of the effusion in some patients. The complication rate was very low; particularly there was no occurrence of infection of any kind. The outcomes and complications of IPC use in ESRD pleural effusions in our series compare favourably with other series of nonmalignant and malignant pleural effusions [Tremblay and Michaud, 2006; Bhatnagar et al. 2014; Srour et al. 2013b].
The current management options for pleural effusions associated with ESRD such as pleurodesis or recurrent thoracentesis have not been compared with IPC and our conclusions are limited by the lack of a control group. The small number of patients also limits generalizability.
There are little data regarding the survival of patients with ESRD and pleural effusions. In one report of patients with peritoneal dialysis, survival was generally poor but varied between subgroups, the median survival in the worst group being about a year [Kwan et al. 2010]. In our previous series of patients treated with IPCs for cardiogenic pleural effusions, median survival was 432 days (95% CI 67–1316) [Srour et al. 2013b]. It is difficult to determine whether survival in the current series is different from what should be expected in these patients and whether IPC insertion had any effect on survival.
Protein store depletion is a theoretical concern with recurring pleural fluid drainage and may predispose to infectious complications. However, our data suggest albumin levels had not changed after a median of about 1 month after the catheter was inserted.
Similarly to previous IPC reports published by us [Srour et al. 2013a, 2013b], the complication rate within this case series was particularly low with only one complication involving a dislodged catheter warranting any further intervention. The series by Bhatnagar and colleagues also had a low complication rate [Bhatnagar et al. 2014]. In contrast, Davies and colleagues reported a serious pleural infection rate of 9.6% [Davies et al. 2012]. The discrepancy may be attributed to the careful selection of patients, the close follow up within the CARE clinic and the trained healthcare workers who are responsible for draining the IPC on a regular basis and troubleshooting as problems arise.
The IPC is particularly useful for cases where invasive treatment options such as pleurodesis, pleural decortication surgery or recurrent thoracentesis would not be in line with the patient’s goals of care. The use of IPCs is limited to centers that have the necessary infrastructure to support the insertion and close follow up required to avoid complications and optimize outcomes. The accessibility of clinic personnel is a central feature in our institution and likely a large part of its success. The unique advantage of using the IPC for patients with ESRD is that these patients are usually followed closely by healthcare workers for hemodialysis or peritoneal dialysis. The dialysis nurses in our institution have been trained to complete regular drainage of the IPC, troubleshoot when issues arrive, and act as a liaison between the patient and the CARE clinic team. This approach has limited the additional expenses, resources and interruptions for the patient that are associated with an IPC and can easily be incorporated in other centers that offer dialysis.
It has been proposed that three groups of patients with malignant pleural effusions benefit most from IPC insertion: (1) patients with the intraoperative finding of a trapped lung in diagnostic video-assisted thoracic surgery (VATS) procedure; (2) patients after a history of repeated pleuracenteses or previously failed attempts at pleurodesis; and (3) patients in a reduced condition with a limited lifespan due to underlying disease [Schneider et al. 2009]. We propose that our criteria may be used to select patients with nonmalignant pleural effusions for IPC insertion as they have been successfully used to identify patients with nonmalignant effusions, both of renal and cardiac origin [Srour et al. 2013b] who have benefited from this procedure. While further research is needed, particularly on the impact of this procedure on quality of life, IPC insertion for selected patients with pleural effusions due to ESRD appears safe and effective.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
The authors declare no conflicts of interest in preparing this article.