Keywords nephrostomy tube exchange - malignant ureteral obstruction - benign ureteral obstruction
Introduction
The insertion of a percutaneous nephrostomy (PCN) catheter was first described by
Goodwin et al in 1955, and it is a common procedure performed by the interventional
radiologists.[1 ] The collecting system of the kidney is accessed through the overlying flank, and
it provides a temporary or permanent alternative drainage of urine. The most common
indication for PCN is urinary obstruction. Nonobstructive conditions including urinary
fistulae, ureteric injury requiring urinary diversion, and percutaneous access to
perform nephrolithotomy are less frequent indications for PCN. The technical aspects
and the periprocedural care of PCN catheters have been well studied.[2 ]
[3 ]
[4 ]
[5 ] However, there are no evidence-based guidelines for long-term management of these
catheters after placement, as literature regarding late complications is scarce.[6 ] The recommended time frame for PCN exchange ranges from 1 to 6 months, based on
institutional preferences. At the institution, PCN exchanges are performed every 3
months irrespective of the underlying disease process. Cancer patients are at an overall
higher risk of developing complication compared with noncancer patients.[7 ] The incidence of PCN catheter-related complications has been reported to be as high
as 19% in cancer patients.[7 ]
[8 ] The objective of this study is to determine the timeline of nephrostomy exchanges
in both cancer and noncancer patients and to determine whether and at what interval
a routine exchange would be recommended.
Materials and Methods
Study Population
Institutional review board (IRB) approval for this study was obtained in October 2014
from University of Texas Southwestern Medical Center, Dallas, Texas, United States,
and this study is in compliance with Health Insurance Portability and Accountability
Act (HIPAA) guidelines for research. All patients who underwent PCN placement/exchange
with at least one subsequent follow-up PCN exchange between January 2011 and January
2014 were identified from the hospital database. Patients who had placement of a PCN
during the study period with no follow-up exchanges, procedures involving other percutaneous
ureteric catheters (e.g., double-J stents, nephroureteral stents), and PCN exchanges
performed within 48 hours of placement/exchange were excluded, as these are typically
due to periprocedural complications.
Data Review
Data of the electronic medical records about the patients who underwent PCN placement/exchange
procedures were reviewed and recorded into a database that included date and reason
for PCN placement/exchange. An exchange was defined as routine or elective if the
patient presented for the scheduled PCN exchange at 90 ± 15 days and did not have
complication at the time of exchange; that is, the patient presented with a normal
functioning PCN without obstruction, infection, or mechanical failure of the tube.
An exchange was defined as emergent or nonelective if the patient presented secondary
to infection, tube obstruction, or mechanical failure. Infections were considered
to be present if the patient presented with pyuria, flank pain, fever, or sepsis associated
with elevated white blood cell count or positive urine/blood cultures. Asymptomatic
bacteria and catheter exit site infections were excluded. Mechanical failure included
catheter dislodgement (partial or complete), malposition, or catheter defects (catheter
being cut). Catheter obstruction was considered to be present if there was no output
from the PCN without signs of infection or evidence of mechanical failure. Catheter
survival day was defined as the time in days between the date of the exchange of interest
and the date of the most recent previous placement or exchange (e.g., if a certain
exchange was the patient's third complication during the study period, the time was
determined from the date of the second complication rather than the date of the initial
placement). Other characteristics recorded in the database included sex, age, race,
reason for initial placement (e.g., cancer, nephrolithiasis, and other less common
reasons for PCN such as retroperitoneal fibrosis). [Table 1 ] shows the distribution of study group characteristics.
Table 1
Patient characteristics (n = 103)
Characteristics
Value
Mean age (y)
48.35
Men/women
36 / 67
Disease process:
Cancer
48
Bladder
4
Cervical
22
Colorectal
7
Prostate
5
Others/metastasis
10
Nephrolithiasis
39
Statistical Analyses
Comparisons of the mean, median, and standard deviation for the number of catheter
days for tube exchange were made within each patient group (e.g., routine vs. obstruction,
routine vs. infection, routine vs. mechanical, etc.) and between the cancer and noncancer
patient groups. Statistical analyses were performed using unpaired Student t -test or one-way analysis of variance (ANOVA) followed by post-hoc Student-Newman-Keuls
test when applicable. p value ≤ 0.05 was considered statistically significant.
Results
Patient Characteristics
Total 103 patients underwent initial PCN placement followed by one or more PCN exchanges
during the study period, accounting for a total of 256 procedures.
Of these, 48 patients had cancer whereas 55 patients did not. Twenty-four (50%) cancer
patients had more than one PCN exchange procedure, whereas 22 (40%) noncancer patients
had more than one PCN exchange. The number of patients who had bilateral PCNs were
14 and 11 in the cancer and non-cancer groups, respectively. The mean age of the cancer
patients was 53.7 years compared with 43.6 in the noncancer cohort. The cancer cohort
had 37 females compared with 30 in the noncancer cohort.
Reason for Catheter Exchanges:
Of the 256 PCN exchanges, 139 were performed in the cancer cohort and 117 in the noncancer
cohort. Routine exchanges accounted for 43% of the cases in cancer patients and 55%
in noncancer patients. Nonroutine exchanges were more common (57% vs. 45%) in the
cancer patients compared with the noncancer cohort, and they accounted for more than
half of the procedures in cancer patients. Among the complications, obstruction was
the most common cause in cancer patients, which accounted for 32% of total procedures,
followed by infection (16%) and mechanical failure (9%). In the noncancer group, the
number of cases of obstruction, infection, and mechanical failures is very similar,
accounting for 17 to 18% of emergent tube exchange procedures.
Catheter Survival Duration
Catheter survival days for routine and nonroutine PCN exchanges are recorded in [Table 2 ]. Comparing the routine exchanges for the cancer and noncancer cohort showed that
the catheters were exchanged at 93 ± 21 and 102 ± 23 days, respectively (not statistically
significant). The catheter survival days in the cancer cohort were 50 ± 32 days for
obstruction, 96 ± 50 days for infection, and 55 ± 23 days for mechanical failure.
The catheter survival days were 103 ± 81, 93 ± 48, and 65 ± 50 days for obstruction,
infection, and mechanical failures, respectively, in the noncancer cohort. A statistically
significant difference was observed when comparing the catheter survival days in the
cancer versus the noncancer cohort only in the PCN obstruction category (p < 0.05). In the cancer cohort, 56% of the nonroutine exchanges were due to obstruction
(n = 45). In comparison, there was not a predominant reason for nonroutine exchange
in noncancer patients (n = 17 for obstruction, n = 18 for infection and mechanical). Mechanical issues in both patient groups tend
to occur earlier than 3 months and have much shorter catheter survival days.
Table 2
Catheter days until PCN exchange
Routine
Obstruction
Infection
Mechanical
Abbreviations: PCN, percutaneous nephrostomy; SD, standard deviation.
a
p < 0.01 between cancer versus noncancer in obstruction group.
b
p < 0.05 versus the respective routine group.
c
p < 0.05 versus the respective obstruction group.
d
p < 0.05 versus the respective infection group.
Cancer
Mean ± SD
102 ± 23
50 ± 32a,b
96 ± 50c
55 ± 23b,d
Median, 25%
96, 90
39, 24
90, 59
51, 42
N
59
45
22
13
Noncancer
Mean ± SD
93 ± 21
103 ± 81
93 ± 48
65 ± 50b,c,d
Median, 25%
91, 90
86, 35
91, 73
53, 24
N
64
17
18
18
[Fig. 1 ] is the distribution of frequency of complications occurring in the cancer and noncancer
cohorts. In the cancer cohort, more than half of those who had obstruction experienced
it within 45 days of tube insertion (median of 39 days), and the rate of mechanical
failure was highest between 45 and 76 days (median of 51 days). In contrast, 64% of
noncancer patients in this study did not require a nonroutine tube exchange before
3 months. The leading cause of early nonroutine tube exchange in this patient population
is mechanical failure such as due to tube dislodgment, malposition, or leakage. In
both groups, only a small proportion of emergent tube exchanges occurred beyond 105
days.
Fig. 1 (A) Comparison of causes of PCN failure in patients with cancer. Obstruction was
observed to have an early peak between 16 and 45 days. Mechanical failures also tend
to occur before the 3-month routine exchange. The highest rate of infection occurred
during 76 to 105 days. (B) Comparison of causes of tube failure in patients without
cancer. The complication risks were less common and relatively more evenly spread
in this group.
Discussion
PCN placement has an overall technical success of 88 to 94% for urinary decompression
with approximately 10% major and minor complication rate combined and 0.2% mortality
rate.[9 ]
[10 ] Postprocedural management of PCN is not standardized, and there are limited data
on the optimal time for exchanges.[11 ]
[12 ] The clinical presentation, definitive treatment, and outcomes are different for
malignant and benign ureteral strictures requiring a PCN.[11 ] Benign ureteral obstruction (BUO) typically presents acutely with pain, nausea,
and vomiting associated with or without signs of pyelonephritis whereas malignant
ureteral obstruction (MUO) is often asymptomatic or may present with uremia or vague
abdominal discomfort. Rarely MUO presents acutely with pyelonephritis/sepsis. The
indication for urinary decompression is for temporary relief before definitive treatment
in BUO whereas PCN placement is more likely to be a definitive measure to preserve
renal function in MUO. Patients with MUO usually have a poor prognosis and lower overall
survival with limited chance for complete cure unlike BUO.[11 ]
This study showed higher complication rates necessitating earlier PCN exchange in
cancer patients compared with noncancer patients. The reasons for high PCN failure
rate early on in cancer patients have not been studied. Monsky et al also reported
a higher rate of complications, including catheter dislodgments, pain, infection,
and clogging in patients receiving PCN for MUO. In their study, 83% required additional
PCN changes within the 3-month period.[12 ] Bahu et al reported only 45% of cancer patients underwent routine catheter exchange
per the institutional policy.[7 ] In their cohort, the most common complication was tube obstruction (24%) followed
by pyelonephritis (19%), within 3 months of the procedure, whereas half of those cases
occurred within the first month. Their findings are similar to ours in which routine
exchanges were done only in 43% of the exchanges and catheter obstruction was the
most common complication accounting for 32% of all exchanges and infection 16%.
In contrast, 55% of patients with BUO underwent routine PCN exchange. Compared with
MUO, patients with BUO are relatively younger with better performance status. At the
institution, patients are referred for a routine exchange by noninterventional radiology
(non-IR) clinics. Although the institutional policy is for a routine exchange at 3
months, this is not strictly adhered to due to variability of clinic follow-up. These
complications are preventable with a strict follow-up program. The authors have initiated
an IR-driven PCN clinic to address this nonuniformity.
There are limitations to this study. The procedural indication, patient education
regarding tube hygiene, the method, and timing of follow-up are not consistently documented.
Patients are referred for PCN exchanges from different specialties with variable follow-up
policies. Many of the patients have routine PCN exchanges and complications beyond
the 3-month mark, which would bias the data. This study excluded all patients who
returned for PCN-related complications within the first 48 hours as the authors defined
these as procedure related. Though the authors are interested in nonprocedural-related
complications, they are fully aware that the 48-hour time limit is arbitrary, and
that there is no strict timeline to separate procedure- versus nonprocedural-related
complications. Last, for those with bilateral PCN, exchanges are typically performed
for both PCNs at the same time, and they were considered two procedures in this analysis.
In conclusion, patients who receive PCN for BUO and MUO are different and require
different follow-up protocols. Patients with MUO develop PCN tube–related complications
earlier than those with BUO. The complication rates are higher in MUO. A universal
3-month nephrostomy tube exchange policy may not be ideal for both the groups. A prospective
study with an established IR-driven PCN clinic is required to define an optimal policy
for postprocedural nephrostomy tube care in both the cohorts.
