Revie

w Journal of the Royal Society of Medicine; 2018, Vol. 111(4) 125–

135
DOI: 10.1177/0141076818766725

Malignant ureteric obstruction decompression: how

much gain for how much pain? A narrative review

Joanna Prentice1, Tarik Amer1,2, Ali Tasleem3 and Omar Aboumarzouk1,2,4

1
Department of Urology, Queen Elizabeth University Hospital Campus, Glasgow G51 4TF, UK
2
Urological Research Unit, Queen Elizabeth University Hospital Campus, Glasgow G51 4TF, UK
3
University College London Hospital, 235 Euston Road, London NW1 2BU, UK
4
College of Medical Veterinary and Life Sciences, Queen Elizabeth University Hospital, Glasgow G51
4TF, UK
Corresponding author: Joanna Prentice. Email: joannaprentice@doctors.org.uk

Summary with this multi-disciplinary approach; in addition, the

Over the last thirty years, the management of Malignant surgical approach to malignant ureteric obstruction
Ureteric Obstruction (MUO) has evolved from a single has evolved from predominantly highly morbid open
disciplinary decision to a multi-disciplinary approach. surgi- cal procedures4 to minimally invasive
Careful consideration must be given to the risks and ben- techniques.5 Brin et al. described their ‘disappointing’
efits of decompression of hydronephrosis for an individual experiences of open palliative procedures with
patient. There is a lack of consensus of opinion as well as patients suffering ‘an inexor- able downhill course’.6
strong evidence to support the decision process. Interestingly, oncologists are more likely to push for
Outcomes that were identified amongst patients under- decompression in asymptomatic patients with a poor
going treatment for MUO included prognosis, quality of
prognosis than urologists.7
life (QOL), complications, morbidity and prognostication
tools. A total of 63 papers were included. Median survival Individualised consideration must be given to the
was 6.4 months in the 53 papers that stated this outcome. risks and benefits of decompression. 6,8–12 Although
Significant predictors to poor outcomes included low there are recommendations within cancer-specific
serum albumin, hyponatremia, the number of malignancy guidelines, both the European Association of Urology
related events, and performance status of 2 or worse on and the American Urological Association guidelines
the European cooperative cancer group. We propose a recommend decompressing the urinary systems,13,14
multi-centre review of outcomes to enable evidence- there is a lack of consensus of opinion as well
based consultations for patients and their families. as
strong evidence to support the decision
Keywords process.2,4,16,17 The National Institute for Health and
Clinical, end of life decisions (palliative care), oncology, pal- Care Excellence guidelines concluded that patients
liative care, urological cancer, urology
should be offered decompression, but that the option
of ‘no intervention should also be discussed’. They
noted that there was insufficient low-grade evidence
in this arena. 16,17 None of these recommendations
take into considera- tion the implications of quality
Introduction of life.
Malignant ureteric obstruction is a condition that To this end, we aimed to conduct a review of the
affects patients with advanced stages of cancer. An literature to be able to inform the decision-making
obstructed single system can significantly reduce process of managing patients with malignant ureteric
patients’ quality of life especially if infection ensues; obstruction. Specifically, we aim to distil the relevant
however, bilateral obstruction will lead to a evidence in this paper to help facilitate an evidence-
certain death. In fact, upper urinary tract based consultation with patients and their families on
obstruction is a prognostic indicator of morbidity prognostic outcomes of decompression in the
for many cancers.1–3 Over the last 30 years, the Methods
setting of malignant ureteric obstruction. 16
management has evolved from a single disciplinary
Search strategy
decision to a multi-disciplin- ary approach involving
urologists, oncologists, pallia- tive care physicians, The review was conducted using Cochrane and
general medicine physicians and interventional PRISMA guidelines.17–19 The search strategy included
radiologists. This is mainly due to the fact that the following databases: the US National Library of
advanced stages of cancer is now treated
A The Royal Society of Medicine 2018
Reprints and permissions: sagepub.co.uk/journalsPermissions.nav
126 Journal of the Royal Society of Medicine
111(4)
Medicine’s life science database (MEDLINE) (1975– who received decompression via percutaneous
September 2017), EMBASE (1975–September 2017), nephrostomy or ureteric stenting;
Cochrane Central Register of Controlled Trials – 2. Quality of life associated with the above;
CENTRAL (in The Cochrane Library – 2017), 3. Major and minor complications;
CINAHL (1975–September 2017), Clinicaltrials.gov, 4. Morbidity defined as hospitalisation
Google Scholar and individual urological journals. post intervention;
Search terms used included: ‘malignant ureteric 5. Effect of decompression on renal function;
obstruction’; ‘percutaneous nephrostomy’; ‘stent’; 6. Prognostication tools in use to predict poor out-
‘quality of life’; and ‘prognosis’. comes from intervention.
Medical Subject Headings (MeSH) phrases
included:

● ((‘‘Stents’’[Mesh])
Data extraction
AND
● (((‘‘Stents’’[Mesh])
‘‘Neoplasms’’[Mesh]) ANDAND ‘‘Ureter’’[Mesh])
‘‘Quality of AND Data of each included study were independently
‘‘Neoplasms’’[Mesh])
Life’’[Mesh] AND ‘‘Quality of Life’’ extracted initially by two authors (JP and TA) after
[Mesh] which a senior author (OA) extracted the data inde-
● (((‘‘Stents’’[Mesh]) AND ‘‘Ureteral Obstruction’’ pendently and cross-checked data extraction to
[Mesh]) AND ‘‘Neoplasms’’[Mesh]) AND ensure quality assurance of data. Data were tabulated
‘‘Quality of Life’’[Mesh] using Microsoft Excel and inbuilt formulae utilised.
● (((‘‘Stents’’[Mesh]) AND ‘‘Ureteral Obstruction’’ The following variables were extracted from each
[Mesh]) AND ‘‘Neoplasms’’[Mesh]) study: number of patients; gender; intervention;
AND ‘‘Prognosis’’[Mesh] age;
● ((‘‘Nephrostomy, AND primary diagnosis; median survival;
Percutaneous’’[Mesh]) Life’’ complications; amount of time spent in hospital;
‘‘Neoplasms’’[Mesh]) AND ‘‘Quality of proportion of
[Mesh] lifetime spent in hospital; proportion of patients
● ‘‘Ureteral Obstruction’’[Mesh]) AND ‘‘Neoplasms’’
(((‘‘Nephrostomy, not
Percutaneous’’[Mesh])
[Mesh]) AND
AND ‘‘Prognosis’’[Mesh] discharged; mortality, prognostication (where
● (((‘‘Nephrostomy, Percutaneous’’[Mesh]) AND
Results
avail- able); and quality of life.
‘‘Ureteral Obstruction’’[Mesh]) The initial review yielded 169 papers (see Figure 1).
‘‘Neoplasms’’[Mesh]) AND AND Of these, 54 were excluded after abstract screening
‘‘Quality Life’’[Mesh] of and 47 were later excluded after full manuscript
review. Of the 47 papers excluded, 18 papers included
benign causes, 16 had no survival data, four
used
Study selection
Three authors (JP, TA and OA) independently com- Figure 1. PRISMA flow chart.
pleted the review of literature independently and fol-
lowed predefined inclusion criteria. Disagreement Initial search n = 160 Hand search n =9
between the authors in study inclusion was resolved
by consensus.

Inclusion criteria Abstract screening Excluded after

n=112 manuscript review n=47:
All types of publications were included. Manuscripts
involving adult patients (18 years old and above) with Benign disease n =18
malignant ureteric obstruction in the English lan- No survival n = 16
guage were included. If only abstracts were available,
these were included if sufficient data were Surgical diversion n = 4
extractable. We included papers reporting on benign
Other n = 9
disease if the data could be extracted separately.
Our outcome measures were:
Included n = 63

1. Prognosis in patients diagnosed with malignant

ureteric obstruction (across all tumour groups)
Prentice et 127
al.
surgical diversion techniques and one paper excluded Eight papers provided a mean one-year survival;
patients with poor outlook. Four were not available the aggregate mean of the percentage of patients
in the English language. Three authors were con- sur- viving one year was 23%. 3,9,10,24,25,29,33,39,45
tacted to obtain manuscripts but did not respond.
In total, 63 papers were included in the review.
Quality of life
Twenty papers assessed quality of life with a total of
Characteristics of included studies
824 patients. 9,11,12,20–23,25,28,31,33,45,46,49,50,61,62,66–68
Seventeen studies were from United States of Measures included time spent in hospital, pain assess-
America, 14 from the United Kingdom, seven from ment and qualitative interviews.
Japan, five from Brazil, three from Germany, two Five studies used the Grabstald outcome measure
from Greece, two from Korea. There was one paper tool (167 patients). 20,21,50,62,67,69 A cumulative analy-
authored from Serbia, the Philippines, Singapore, sis which found that 60% of patients were able to
China, Pakistan, Jordan, Turkey, Israel, Sweden, achieve a ‘useful life’ post decompression (Table 2).
New Zealand, Australia and Austria. Two studies used the Functional Assessment
Only nine studies were prospective in nature; of of
these, one was a prospective cohort study. There Cancer Therapy concerning nephrostomy insertion
were no randomised controlled trials. The follow-up and nephrostomy vs. stent insertion (Table 3).
period ranged from six months to eight years. Neither had a significant difference between the
groups. One study used the European Organization
for Research and Treatment of Cancer Quality of
Demographics Life Questionnaire. 61 There was no significant differ-
In total, 4948 patients were included ence in quality of life when administered pre- and
in this study. 1 , 3 , 8 – 11 , 1 6 , 2 0 – 4 4 , 4 4 – 6 3 Of these, 1030 post-nephrostomy insertion (Table 3) Aravantinos
had stents and 3891 had nephrostomies. Most papers
patients et al. used the European Organisation for Research
classified patients by individual tumour type and Treatment of Cancer Quality of Life
(Table Questionnaire. 61 There was no significant difference
1). 1 , 3 , 8 – 11 , 1 6 , 2 0 – 4 4 , 4 4 – 6 4 The mean age of patients was 60 in quality of life when administered pre and post
nephrostomy insertion.
years (range: 19–97 years). 1,3,8–11,16,20–26,28,29,32,33,35–
39,41–43,45,48–50,52–54,56–58,60,61,63,64
Complications
Prognosis Twenty-four of the papers commented on the
Fifty papers included prognosis as an outcome mea- frequency of complications with a total of 1891
sure with a total of 2790 patients included. This patients. The overall complication rate was
ranged from 21 h to 140 months with a median sur- 41%. 10,11,16,22,23,26,32,33,36,39,45,51,59
1,3,8–11,16,20–44,44–53,55,57–60,62–65
vival of 6.4 months. Twenty-six per cent (439/1658) of patients with
nephrostomies developed urinary infection, while 14%
(26/180) of patients with stents placed developed infec-
tions. Ten per cent (173/1658) of patients experiencing
Table 1. Distribution of dislodged nephrostomies, while 7% (113/1658) of
cancers. patients developed blocked nephrostomies. Stent migra-
No. of patients
Type of cancer included tion/dislodgement was reported in 6% (10/180).
Haematuria rate was 8% (15/180) in patients stented
Prostate 1561 compared to 3% (49/1658) in patients with nephros-
tomies. Nephrectomy rate was 0.2% (4/1658) following
Cervical 829
percutaneous nephrostomy placement; two for peri-
Bladder 533 nephric abscess (the indication for the other two
patients was not stated).32,33 Mortality rate was 0.2%
Colorectal 473 (4/1658): three from haemorrhage and one from sepsis.
In the three papers that reported mortality, the overall
Gastrointestinal 300
rate was 5% (4/82).22,52,62
Uterine 64 Twelve papers (628 patients) calculated the pro-
portion of patients who never left hospital
Other 605 decompression, 9,11,21–23,28,31,33,45,55,62,64 with post
the
pooled mean for this being 26% (range: 5–69%).
128 Journal of the Royal Society of Medicine
111(4)
Table 2. Papers using Grabstald ‘useful life
measure’.
Number and
male/female Stent/ Grabstald
Paper Type Date average age Tumour type nephrostomy (N percentag
patients) e
Hubner et al.50 Retrospective 1986–1989 52 Prostate 7% Stent 24, PCN 28 81%
(31 F, 21 M) Bladder 25%
67 (43–81) Colorectal 28%
Cervix 17%
Ovarian 11%
Other 2%

Hoe et al.62 Retrospective Not stated Not stated Colorectal 33% PCN 24 46%
Cervix 5%
Prostate 5%
Bladder 5%
Rest not stated

Emmert et al.20 Retrospective 1990–1995 24 Cervical 100% PCN 24 46%

45.9 (30–79)

Feng et al.67 Retrospective 1984–1996 37 Prostate 27% Stent 22, PCN 15 82–87%
(20 F, 17 M) 37– Bladder 13% classified into
85
No mean Colorectal 10% two groups
Cervix 32%
Uterus 5%
Ovarian 10%

Wilson et al.21 Retrospective 1996–2001 32 (16 M, Prostate 28% PCN 32 46.9%

16 F) Bladder 25%
68.1 (42–84) Colorectal 21%
Cervix 15%
Uterus 6%
Breast 3%

Patients spent 20% of their remaining lifetime in hos- Patients with a malignancy of unknown primary
pital. 8,23,28,33,37,44,45,48,49,53,62,64 Twelve papers or gastrointestinal origin were identified as having
included renal function pre and post procedure (a poorer outcomes, 28,42 whereas gynaecological malig-
total of 1135 patients). Pre-nephrostomy, the average nancies had a better outcome. 38 Other variables
creatinine was 624 mmol/L and post procedure, included patients with upper ureteric obstruction, 24
the creatinine improved to 212 mmol/L on moderate–severe hydronephrosis, 25 bilateral hydro-
average. 9,10,21,23,31,43,44,49,54,60,61,70 nephrosis, 42 elevated creatinine, 25,38 anaemia 56 and
patients with an elevated C-reactive protein. 30
Prognostication tools
Discussion
Sixteen papers with a total of 2061 patients investigated
various factors and their ability to prognosti- This review of 63 papers gives a broad survival range
cate 3,9,10,24,25,28,30,31,36,38,42,54,61,71–73 in these patients for patients with malignant ureteric obstruction
between 21 hours and 140 months. 1 , 3 , 8 – 11 , 1 6 , 2 0 – 4 4 , 4 4 –
(Table 4). 53,55,57–60,62–64
The median survival was 6.4 months
Most commonly occurring statistical significance and the percentage of patients alive at one year was
included low serum albumin, 1 0 , 3 0 , 3 6 , 4 2 , 7 3 no further 23%. 3 , 9 , 1 0 , 2 4 , 2 5 , 2 9 , 3 3 , 3 9 , 4 5 Reasons for this variation in
treatment options,3,24,38,54 hyponatraemia,30,36 number survival include the heterogeneous patient and cancer
of malignancy-related
static disease, events
10,30,36,42
ascites), theeffpresence
(pleural usion, meta-
of groups involved. Additionally, the data are limited
metastatic disease,28,61 performance status of 2 or by the fact that researchers in some instances may
worse on the European Cooperative have included patients with retroperitoneal fibrosis
Cancer Group. 9,24,54,73 second- ary to treatment (such as radiotherapy)
rather than
Prentice et 129
al.
Table 3. Quality of life.
Author Patient details Assessment Outcome

Aravantinos et al.61 207 patients EORTC- No significant difference in QOL when

Bladder, prostate, cervical, QOLC –C30 administered pre and post
gynaecological nephrostomy insertion.

Monsky et al.66 46 patients (13 lost to follow-up) FACT-BL No statistical differences in patients’
Bladder 14, cervical 15, prostate responses post stenting or percutaneous
6, uterine 5, Other 7 nephrostomy insertion. Patients with
PCN ¼ 15 stents reported significantly greater pain
Stent ¼ 31 and storage lower urinary tract
symptoms, although this did not
translate into a reduction in measured
QOL.

Lapitan et al.25 198 patients Cervical cancer FACT-G There was no statistical difference in the
FACT-G scores for patients with or
with- out percutaneous nephrostomy

Bigum et al.12 10 patients (prostate 8, bladder Qualitativ Main themes:

cancer 2) e Lack of follow-up, complications, physical
All nephrostomy interview limitations and the impact on their
social
life
Kumar et al.68 17 patients Qualitativ Main themes:
All percutaneous nephrostomy e Symptoms from decompression, an edu-
Ovary 6, uterine 3, cervical 2 interview cational void and the role of self
education (30% no symptoms)
PCN: percutaneous nephrostomy; QOL: quality of
life.

ongoing or recurrent disease. 74,75 The benign nature present. 71 Despite extensive retrospective publica-
of this aetiology for obstruction would skew results tions and several review articles, there is a paucity
towards improved outcomes. The majority of the of data assessing the important issues of quality of
papers reported on patients who underwent decom- life and prognostication in this cohort of patients. 5
pression, thereby not capturing a proportion of European Association of Urology guidelines on
patients who were not decompressed. pain management recommend that for pelvic malig-
Two pertinent questions are always presented: nancies ‘it is good practice to drain symptomatic
what are the preferred options for relieving malignant hydronephrosis at once, and to drain only one
ureteric obstruction? And what is the expected kidney (the less dilated and better appearing kidney
prog- nosis?5 In terms of methods of decompression, or the one with the better function, if known)
a pre- in
vious comparative study discovered no asymptomatic patients’.14 They conclude that a
relative nephrostomy tube is superior to a double-J stent for
superiority of retrograde stenting to percutaneous drainage for pelvic malignancies but advocate either
nephrostomy in the setting of infected obstructed stenting or nephrostomies in other tumour groups. 14
uropathy caused by stones. 76 Two recent review Neither of these recommendations reference the lit-
articles
my concluded antegrade
± subsequent that there stent
were when
no data on the erature nor do they mention implications of quality
malignant
superiority and
considering benign
of stent vs. ureteric obstruction.
percutaneous nephrosto-
5,77
of life.
Clearly, the aim of relieving the obstruction In the context of locally advanced non-
depends on patient factors but would include improv- metastatic
ing renal function to enable further oncological treat- bladder cancer with hydronephrosis, American
ment, to correct the symptoms of renal failure and Urological Association guidelines suggest placement
to improve pain. 7 This must be balanced against a of a ureteral stent. 13
Complications
patient’s expectation of quantity and quality of life.
This is, of course, a challenging consultation, parti- Another frequently neglected statistic for patients
cularly in the acute setting when such patients often before them undergoing decompression is the
 0
13
Table 4. Summary of literature on
prognostication.
Features of poor
Tumour type outcome (statistically
Paper Study type N Age (range) Process (%) significant) Survival based on predictors

Feuer et al.54 Prospective 22 58 (n/a) Univariate, Cervical 77 Patients having one of: Survival
Kaplan- Gynaecologic progressive tumour, per- 242 days (if 0 factors) vs. 37 days
Meyer al (other) 18 formance status > 2, (if 1 or more factor)
Uterine 4 tumour-related medical Days at home
problems, no treatment, 164 days (if 0 factors) vs. 37 days
uncontrolled pain (if 1 or more factor)

Watkinson Retrospective 50 53 Grouping – no Cervical 32 Not identified 12-month survival

et al.72 case series statistical analysis Bladder 36 Group I – benign or treated (100%)
Colon 10 Group II untreated malignancy (50%)
Lymphoma 4 Median survival 339 days
Ovary 4 Group III ureteric obstruction by
Other 6 abdominopelvic disease with treat-
ment (50%)
Median survival 334 days
Group IV ureteric obstruction by
abdominopelvic disease without
treatment (0%)
Median survival 38 days

Wong et al.28 Retrospective 102 62 (31–86) Univariate and Gastric 20 Presence of metastatic 12-month survival
case series multivariate analy- Gynaecological 31 disease 63% in favourable (0–1 unfavourable
sis Kaplan-Meyer Urological 29 Diagnosis of MUO in factors)
Other 18 presence established 12% in unfavourable (4 unfavourable

Medicine 111(4)
Journal of the Royal Society of
malignancy factors)

Jeong et al.24 Retrospective 86 54 Univariate and Gastric 33, ECOG 2 or more No prognostication tool suggested
case series (23–79) multivariate Cervical 10 Lesion in upper ureter
analysis Colorectal 40, No treatment options
Prostate 0, post nephrostomy
Bladder 0, Other
17

Aravantinos Retrospective 507 63 (40–85) Kaplan-Meyer Bladder 32 Disseminated disease All patients six-month survival
et al.61 curves and Mann- Prostate 26 (significant for prostate 33% survival > 6 months
Whitney for QOL Colorectal 17 and colorectal cancer
Gynaecological 13 only)
Other 8 gastric/
pancreatic 3

(continued)
 al.
Prentice et
Table 4. Continued.
Features of poor
Tumour outcome (statistically
Paper Study type N Age (range) Process type (%) significant) Survival based on predictors

Ishioka et al.10 Prospective 140 57 (31–85) Multivariate analy- Gastric 21, 3 or more cancer-related Six-month survival
sis Kaplan-Meyer Cervical 21 events 69% in favourable group (0 risk
Urothelial 9, Low serum albumin fac- tors)
Colorectal 24, Low-grade Intermediate group 24% (1 risk
Other 10 hydronephrosis factor)
Poor group 2% (2–3 risk
factors)
Lienert et al.36 Retrospective 49 71 (36–91) Univariate Cervical 6 Three or more cancer- Mean survival in months
Colorectal 12 related events 9 months for 0 risk factors
Bladder 36, Low serum albumin 5.7 months for1 risk factor
Prostate 30 Low serum sodium 2 Months for 2 or 3 risk factors
Other 14

Jalbani et al.31 Prospective 40 No average Paired T Cervical 37 None identified Good prognostic features
(21–70) Rectum 7 Recent diagnosis
Bladder 25, pros- Age < 52
tate 12 Rectum 7

Izumi et al.38 Retrospective 61 64 Univariate and Gastric 24 High creatinine Survival in months
(27–89) multivariate Cervical 26 (>106 Good predictors 13.2 months
analysis Colorectal 12, mmol/L) Intermediate 8.2 months
ovarian 9 No treatment options Poor 1.7 months
Bladder 3 Non-gynaecological
Prostate 5 cancer

Migita et al.3 Retrospective 25 61 Multivariate analy- Gastric 100 No treatment options Survival
sis Kaplan-Meyer 3.1 months if no treatment options
11.2 months if treatment options

Azuma et al.30 Retrospective 214 79% < 80 Multivariate Gastric 16 Number event related to Median survival
21% > 80 Cervical 15 malignant dissemination 12 months, Favourable (0–1 risk fac-
Colorectal 21 Low serum albumin tors) 6 months, Intermediate (2 risk
Urothelial 15 Low serum sodium factors) 2.6 months, Poor group
Prostate 14 High CRP (3 risk factors)
Other 28

Souza et al.56 Retrospective 48 59 Univariate Cervical 100 Haemoglobin < 8.7 g/dL Not calculated
(6–85) Multivariat Haematocrit < 27% Compared characteristics of group
e analysis Hypotension (in absence who died compared with survival
of sepsis) group

(continued)

1
13
 2
13
Table 4. Continued.
Features of poor
Tumour type outcome (statistically
Paper Study type N Age (range) Process (%) significant) Survival based on predictors

Alawne Retrospective 211 59 Multivariate analy- Gastrointestin Presence of ascites, 12-month survival
h et (6–85) sis Kaplan-Meyer al 28 pleural effusion, 0 factors 78%, 1 factor 36%
al.42 Genitourinary low serum albumin, 2 factors 17%, 3 factors 6%
58 bilateral hydronephrosis
Other 13 gastrointestinal
malignancy

Cordeiro Prospective 208 61 Multivariate analy- Cervical 20 Number malignant 12-month survival
et al.9 (19–89) sis Kaplan-Meyer Colorectal 21 events > or equal to 4 Favourable (0 factors) 45%
Bladder 22 ECOG > or ¼ 2 Intermediate (1 factor) 15%
Prostate 12 Unfavourable (2 factors) 7%

Downe Retrospective 86 – Grouping – no Bladder 35 Not identified 12-month survival

y et statistical analysis No further data (based on Watkinson et
al.71 al.72)
(abstrac Group I (non-malignant) no data
t only) Group II (untreated primary) 20%,
Group III (relapsed disease viable

Medicine 111(4)
Journal of the Royal Society of
option) 16%
Group IV (no treatment option)
0%
Gandiya Retrospective 193 70 Multivariate analy- Urological 47, Low serum albumin Pooled survival
et al.73 (26–90) sis Kaplan-Meyer gynaecological 22 ECOG > or ¼ 2 82% survival at one month
(abstract only) Colorectal 11, Prior oncological 63% at three months
other 20 treatment 50% survival at six months
ECOG: European Cooperative Cancer Group; MUO: malignant ureteric obstruction; QOL: quality of
life.
Prentice et 133
al.
Table 5. Proposed prognostication tool symptoms and not having a control group. 66 Lapitan
(PALLIATE). et al., despite conducting a prospective study
Performance status (ECOG2) using functional assessment of cancer therapy –
Albumin (low) bladder, demonstrated lower scores pre and post
decompres- sion in comparison to other papers
Low serum sodium using functional assessment of cancer therapy –
general.25 This is potentially related to the
Laterality
association between socio- economic deprivation,
Inflammatory markers (CRP) educational attainment and scores in functional
assessment of cancer therapy – general. 80,81 A
Ascites further limitation of the Lapitan et al. study,
Tumour type reducing its wider relevance, was the inclusion of
patients solely with a diagnosis of cervical cancer.
Events related to cancer (pleural effusions, metastatic Two papers looked at qualitative interviews.
disease) Qualitative analysis is helpful to develop themes.
However, a small sample size, the challenges of con-
founders, single tumour group inclusion and the
proportion of time spent in hospital and the risk of exclusion of patients without nephrostomies may
complications. The proportion of patients who had limit its wider application. 12,68
complications was 41%, with 26% of patients never
Prognostication
leaving hospital. Those who had an intervention
spent 20% of their resultant lifetime in hospital. One group performed a prospective cohort study of
One paper stated that 69% of their patients never patients with cervical cancer. Lapitan et al. followed
left hospital. Removing this apparent outlier from up a cohort of patients who had malignant ureteric
the pooled mean resulted in the figure of 17.8%. 22 obstruction and assessed the outcomes of two groups:
Despite improving renal function, creatinine did not those who were decompressed and those who were
return to baseline for patients, potentially avoiding not. 2 5 At the outset, there appears to be a survival
an emergency situation but not reversing the benefit with 38% vs. 28% survival at six months for
damage caused by hydronephrosis. those who underwent decompression vs. those who
did not. By 12 months, however, both groups had the
same survival of 16%. 2 5
Quality of life The most frequently found statistically significant
Quality of life can be challenging to measure; early indicators of poor prognosis among the literature
papers measured quality of life using the ‘useful life’ were low serum albumin, no further treatment
measure. 50,67 Feng et al. and Hubner et al. reported options, number of malignancy-related events
greater proportions of patients achieving a good (pleural effu- sion, metastatic disease, ascites),
quality of life when compared with contemporary performance status of 2 or worse on the European
papers: 81–87% vs. 46%. 2 0 , 2 1 , 5 0 , 6 2 , 6 7 While it is not Cooperative Cancer Group, the presence of
entirely clear why there has been a reduction in metastatic disease and hypo- natremia (Table 4).
patients experiencing ‘useful life’, it is possible that Two papers divided patients into groups depending
patient selection, subjective clinician perception of on treatment options available; those with no
pain and possibly the increased use of opioid medica- treatment options had 0% 12-month survival and a
tion may have contributed. 78,79 Furthermore, it is median survival of 38 days. Combining these
likely that the progression to use of patient-reported parameters with a larger patient group may help
outcome measures rather than relying on clinicians’ develop a prognostication tool for clinicians to aid
opinions of what constitutes quality of life can decision-making (See Table 5).
explain in part the move from rudimentary to more Comparison of prognostication tools demonstrates
patient-centred validated measures. that patients with none or one risk factor have more
Three studies utilised patient-reported outcome favourable outcomes with 12-month survival ranging
measures demonstrating no statistically significant from 20% to 78%28,42,71,72 and a median survival ran-
improvements in quality of life pre and post decom- ging between 9 and 13 months. 30,36,38 In those patients
pression. 25,61,66 However, despite improvement in the with ‘intermediate’ risk factors (see Table 3), median
use of patient-reported outcome measures, there were survival ranged from 5.7 to 8.2 months. 30,36,38 For
several limitations with the studies. Limitations of those patients with two or more risk factors, median
the Monskey et al. study included not measuring survival ranged from 1.7 to 2.6 months30,38 and 12-
baseline month survival ranged from 0% to 12%. 9,28,42,71,72
134 Journal of the Royal Society of Medicine
111(4)
Limitations Ethical approval: Not applicable.
The reviewed data have significant heterogeneity, Guarantor: JP.
making conclusions regarding specific tumour types
Contributorship: JP and TA developed the idea for the article.
challenging. Sountoulides et al. comments on the JP, TA and OA undertook data collection. OA, TA, JP and AT
‘divergent nature of their group’. At one end of the undertook analysis of the data and interpretation. JP and TA
spectrum, there are patients with very advanced dis- drafted the article. OA and AT provided critical revision. JP,
ease who may not benefit from decompression; at the TA, OA and AT involved in final approval.
other are those who have further oncological and Provenance: Not commissioned; peer-reviewed by Syed
or surgical options who will naturally have a much Shahzad.
longer life expectancy.5
Acknowledgements: The authors would like to thank the
There are only 25 papers published in the last 10 library staff at Ayr hospital for their help with sourcing
years; in this time, however, there have been signifi- papers for the review.
cant advances in oncological treatment options.
Therefore, outcomes may well be influenced by References
more conservative historic data. (Refs. 31–82 are available in the online supplemental
Another limitation of this analysis is the retrospec- material)
tive nature of the data; there were only nine papers 1. Radecka E, Magnusson M and Magnusson A. Survival
that were prospective in nature. 9 , 1 0 , 2 5 , 3 1 , 4 3 , 4 8 , 5 4 Data time and period of catheterization in patients treated
on survival are lacking on those patients conserva- with percutaneous nephrostomy for urinary obstruc-
tively managed with only one paper including tion due to malignancy. Acta Radiol 2006; 47: 328–331.
untreated patients in its survival data, which may 2. Pradhan TS, Duan H, Katsoulakis E, Salame G, Lee
Y-C and Abulafia O. Hydronephrosis as a prognostic
lead to overstating median survival.25 There are
indicator of survival in advanced cervix cancer. Int J
clear worldwide variations in practice regarding dis- Gynecol Cancer 2011; 21: 1091–1096.
charge home. In one series, 69% of patients did not 3. Migita K, Watanabe A, Samma S, Ohyama T,
leave hospital after decompression;28 this contrasts Ishikawa H and Kagebayashi Y. Clinical outcome
with an average of 17% among other papers. and management of ureteral obstruction secondary to
Limitations when reviewing prognostic tools gastric cancer. World J Surg 2011; 35: 1035–1041.
include the fact that the majority of studies were ret- 4. Meyer JE, Yatsuhashi M and Green THJ. Palliative
rospective in nature. The use of statistical analyses urinary diversion in patients with advanced pelvic
included univariate, paired-t as well as multivariate malignancy. Cancer 1980; 45: 2698–2701.
analysis, thus limiting transferability and utility on an 5. Sountoulides P, Pardalidis N and Sofikitis N.
individual patient basis. Endourologic management of malignant ureteral
obstruction: indications, results, and quality-of-life
issues. J Endourol 2010; 24: 129–142.
Conclusion 6. Brin EN, Schiff MJ and Weiss RM. Palliative urinary
diversion for pelvic malignancy. J Urol 1975; 113:
In this post Montgomery era with the concept of the 619–622.
‘reasonable patient’, can we continue to justify discuss- 7. Hyams ES and Shah O. Malignant extrinsic ureteral
ing decompression without stating to patients the evi- obstruction: a survey of urologists and medical oncolo-
dence-based risks and benefits from the emergent gists regarding treatment patterns and preferences.
body of literature?82 An overall complication risk of Urology 2008; 72: 51–56.
41% and up to one-quarter of patients’ remaining 8. Paul AB, Love C and Chisholm GD. The management
lifetime spent in hospital with a median survival of of bilateral ureteric obstruction and renal failure in
6.4 months may encourage clinicians and patients to advanced prostate cancer. Br J Urol 1994; 74: 642–645.
rethink the appropriateness of such interventions. We 9. Cordeiro MD, Coelho RF, Chade DC, Pessoa RR,
Chaib MS, Colombo-Junior JR, et al. A prognostic
propose a contemporary multicentre prospective
model for survival after palliative urinary diversion
review of out- comes of this cohort of patients to for malignant ureteric obstruction: a prospective
enable evidence- based consultations for patients and study of 208 patients. BJU Int 2016; 117: 266–271.
their families. Further work in the domain of 10. Ishioka J, Kageyama Y, Inoue M, Higashi Y and
prognostication is needed to help best identify those Kihara K. Prognostic model for predicting survival
patients who may benefit the most from after palliative urinary diversion for ureteral obstruc-
decompression. tion: analysis of 140 cases. J Urol 2008; 180: 618–621;
Declarations discussion 621.
Competing Interests: None declared. 11. Dienstmann R, da Silva Pinto C, Pereira MT, Small IA
and Ferreira CG. Palliative percutaneous
Funding: None declared. nephrostomy in recurrent cervical cancer: a
retrospective analysis of
Prentice et 135
al.
50 consecutive cases. J Pain Symptom Manage 2008; ureteric obstruction in patients with terminal cervical
36: 185–190. cancer. Arch Gynecol Obstet 1997; 259: 147–151.
12. Bigum LH, Spielmann ME, Juhl G and Rasmussen A. 21. Wilson JR, Urwin GH and Stower MJ. The role of
A qualitative study exploring male cancer patients’ percutaneous nephrostomy in malignant ureteric
experiences with percutaneous nephrostomy. Scand J obstruction. Ann R Coll Surg Engl 2005; 87: 21–24.
Urol 2015; 49: 162–168. 22. Tanaka T, Yanase M and Takatsuka K. Clinical course
13. Chang S, Bochner B, Chou R, Dreocer R, Kamat A, in patients with percutaneous nephrostomy for hydro-
Lerner S, et al. Treatment of non-metastatic muscle- nephrosis associated with advanced cancer. Hinyokika
invasive bladder cancer: AUA/ASCO/ASTRO/SUO Kiyo 2004; 50: 457–462.
guideline. Linthicum, MD: American Urological 23. Shekarriz B, Shekarriz H, Upadhyay J, Banerjee M,
Association: 9. See http://www.auanet.org/guidelines/ Becker H, Pontes JE, et al. Outcome of palliative
muscle-invasive-bladder-cancer-new-(2017) urin- ary diversion in the treatment of advanced
(last checked 8 February 2018). malignan- cies. Cancer 1999; 85: 998–1003.
14. Paez Borda A, Charnay-Sonnek F, Fonteyne V and 24. Jeong IG, Han KS, Joung JY, Seo HK and Chung J.
The outcome with ureteric stents for managing non-
Papaioannou EG. Guidelines on pain management &
urological malignant ureteric obstruction. BJU Int
palliative care. Arnhem: European Association of
2007; 100: 1288–1291.
Urology: 12. See https://uroweb.org/wp-content/
25. Lapitan MCM and Buckley BS. Impact of palliative
uploads/25-Pain-Management_LR.pdf (2014, last
urinary diversion by percutaneous nephrostomy drain-
checked 8 February 2018).
age and ureteral stenting among patients with
15. National Institute for Health and Care Excellence
advanced cervical cancer and obstructive uropathy: a
(NICE). Bladder cancer: diagnosis and management.
prospective cohort. J Obstet Gynaecol Res 2011; 37:
London: NICE; 2015: 385. See https://www.nice.org.
1061–1070.
uk/guidance/ng2/evidence/full-guideline-pdf-3744112 26. Mann WJ, Hatch KD, Taylor PT, Partridge EM, Orr
JW and Shingleton HM. The role of percutaneous
(last checked 8 February 2018).
nephrostomy in gynecologic oncology. Gynecol Oncol
16. Kamiyama Y, Matsuura S, Kato M, Abe Y, Takyu S,
1983; 16: 393–399.
Yoshikawa K, et al. Stent failure in the management of
27. Dudley BS, Gershenson DM, Kavanagh JJ, Copeland
malignant extrinsic ureteral obstruction: risk factors.
LJ, Carrasco CH and Rutledge FN. Percutaneous
Int J Urol 2011; 18: 379–382.
nephrostomy catheter use in gynecologic malignancy:
17. Liberati A, Altman DG, Tetzlaff J, Mulrow C,
M.D. Anderson Hospital experience. Gynecol Oncol
Gotzsche PC, Ioannidis JPA, et al. The PRISMA state- 1986; 24: 273–278.
ment of
lyses for studies
reportingthat
systematic reviewshealth
evaluate and meta-
care 28. Wong L-M, Cleeve LK, Milner AD and Pitman AG.
ana-
interventions: explanation and elaboration. Ann Malignant ureteral obstruction: outcomes after inter-
Intern Med 2009; 151: W65–W94. vention Have things changed? J Urol 2007; 178: 178–
18. Higgins JPT, Altman DG, Gotzsche PC, Juni P, Moher 183; discussion 183.
D, Oxman AD, et al. The Cochrane Collaboration’s 29. Tekin MI, Aytekin C, Aygun C, PeSkircioglu L,
tool for assessing risk of bias in randomised trials. Boyvat F and Ozkardes H. Covered metallic
BMJ 2011; 343: d5928. ureteral stent in the management of malignant ureteral
19. Higgins JPT, Thompson SG, Deeks JJ and Altman obstruction: preliminary results. Urology 2001; 58:
DG. Measuring inconsistency in meta-analyses. BMJ 919–923.
2003; 327: 557–560. 30. Azuma T, Nagase Y and Oshi M. Prognostic marker
20. Emmert C, Rassler J and Kohler U. Survival and for patients with malignant ureter obstruction. Clin
quality Genitourin Cancer 2013; 11: 353–356.
of life after percutaneous nephrostomy for malignant