Study Identification
The initial database search retrieved 4343 articles. After removal of duplicates, abstract review, full text review, and application of inclusion and exclusion criteria, a total of 27 studies, published between 1986 and 2020, were identified. Screening resulted in a total of 26 studies for analysis [10–35]; one additional article from Donnellan et al. (1994) [36] was included after being identified through hand searching the references of included studies. Fig. 1 summarizes the search in a PRISMA flow diagram. The κ was 0.78 (95% CI, 0.70-0.86), indicating strong interrater agreement.
Study Characteristics
Study characteristics, study demographics and urologic history of included patients are summarized in Table 1. Among the 27 included papers, 10 studies included patients undergoing SWL [11–13, 15–19, 27, 36], 17 studies included PCNL [10, 14, 15, 19–25, 27, 28, 30–32, 35, 36], and seven studies included ureteroscopy [19, 26, 27, 29, 33, 34, 36]. Four studies included multiple treatment modalities [15, 19, 27, 36]. The included articles consisted of four studies comparing PCNL between SCI and non-SCI patients [14, 28, 31, 35], and 23 non-comparative studies examining all three interventions [10–13, 15–27, 29, 30, 32–34, 36]. All studies incorporated non-randomized observational designs and were reported retrospectively. Patient data extracted from included studies were from 1986-2018. The mean MINORS scores were 9.4 (± 1.2) for non-comparative studies and 18 (± 1.6) for comparative studies, indicating moderate risk of bias for both study types (Supplemental Table S1).
Table 1
Study characteristics, patient demographics, and urologic history of patients with spinal cord injury and nephrolithiasis
|
|
|
|
|
SCI etiology
|
|
Bladder management
|
|
|
|
Study
|
Patients
|
Average age (years)
|
Male (%)
|
Traumatic (%)
|
Spina bifida (%)
|
Other (%)
|
|
Indwelling (%)
|
SPC (%)
|
CIC (%)
|
Condom (%)
|
None (%)
|
Urinary diversion (%)
|
Bladder augmentation (%)
|
Bacteriuria (%)
|
Culkin et al. (1986)10
|
23
|
43
|
100
|
NR
|
NR
|
NR
|
|
60.9
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
100
|
Lazare et al. (1988)11
|
32
|
52
|
100
|
93.7
|
0
|
6.3
|
|
28.1
|
15.7
|
0
|
46.9
|
0
|
9.4
|
NR
|
93.8
|
Spirnak et al. (1988)12
|
5
|
39
|
100
|
100
|
0
|
0
|
|
40.0
|
0
|
20.0
|
20.0
|
0
|
20.0
|
NR
|
100
|
Wahle et al. (1988)13
|
31
|
34
|
74
|
54.8
|
29.0
|
16.1
|
|
NR
|
19.4
|
NR
|
NR
|
NR
|
41.9
|
NR
|
NR
|
Culkin et al. (1990)14
|
35
|
NR
|
84
|
NR
|
NR
|
NR
|
|
71.4
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
100
|
Irwin et al. (1991)15
|
16
|
NR
|
NR
|
31.2
|
68.8
|
0
|
|
NR
|
NR
|
NR
|
NR
|
NR
|
50.0
|
NR
|
18.8
|
Niedrach et al. (1991)16
|
11
|
31
|
NR
|
36.4
|
36.4
|
27.2
|
|
18.2
|
0
|
9.1
|
45.4
|
0
|
27.2
|
NR
|
90.9
|
Deliveliotis et al. (1994)17
|
15
|
39
|
NR
|
40.0
|
33.3
|
26.7
|
|
73.3
|
0
|
26.7
|
0
|
0
|
NR
|
NR
|
NR
|
Robert et al. (1995)18
|
15
|
27
|
87
|
93.3
|
0
|
6.7
|
|
6.7
|
NR
|
40.0
|
NR
|
13.3
|
NR
|
6.7
|
33.3
|
Donnellan et al. (1999)36
|
58
|
37
|
88
|
91.4
|
0
|
8.6
|
|
48.6
|
NR
|
1.4
|
NR
|
15.3
|
4.9
|
0.7
|
NR
|
Chen et al. (2002)19
|
77
|
50
|
86
|
NR
|
NR
|
NR
|
|
29.0
|
0
|
6.7
|
56.5
|
1.6
|
NR
|
NR
|
NR
|
Rubenstein et al. (2004)20
|
23
|
44
|
57
|
34.8
|
39.1
|
17.4
|
|
NR
|
NR
|
30.4
|
NR
|
NR
|
34.8
|
4.3
|
91.3
|
Lawrentschuk et al. (2005)21
|
26
|
43
|
88
|
NR
|
NR
|
NR
|
|
42.3
|
NR
|
23.1
|
42.3
|
NR
|
7.7
|
NR
|
84.6
|
Symons et al. (2006)22
|
29
|
39
|
41
|
31.0
|
34.5
|
20.7
|
|
20.7
|
24.1
|
20.7
|
NR
|
6.9
|
27.6
|
3.4
|
NR
|
Knox et al. (2012)23
|
47
|
28
|
NR
|
55.3
|
34.0
|
10.6
|
|
12.8
|
17.0
|
19.1
|
10.6
|
6.4
|
27.7
|
8.5
|
93.2
|
Nabbout et al. (2012)24
|
21
|
53
|
62
|
66.7
|
33.3
|
0
|
|
52.4
|
0
|
38.1
|
0
|
0
|
9.5
|
NR
|
90.5
|
Alsinnawi et al. (2013)25
|
5
|
44
|
60
|
0
|
100
|
0
|
|
20.0
|
0.0
|
80.0
|
0.0
|
0.0
|
NR
|
40.0
|
100
|
Wolfe et al. (2013)26
|
29
|
47
|
100
|
NR
|
NR
|
NR
|
|
67.2
|
17.9
|
7.5
|
0
|
0
|
4.5
|
NR
|
NR
|
Clifton et al. (2014)27
|
95
|
51
|
67
|
58.9
|
15.8
|
25.2
|
|
29.5
|
NR
|
15.8
|
NR
|
NR
|
24.2
|
NR
|
NR
|
Danawala et al. (2015)28
|
31
|
56
|
61
|
NR
|
NR
|
NR
|
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
34.1
|
Tepeler et al. (2015)29
|
19
|
45
|
58
|
52.6
|
0
|
42.1
|
|
15.8
|
10.5
|
31.6
|
15.8
|
5.3
|
NR
|
NR
|
NR
|
Sofimajidpour et al. (2016)30
|
29
|
49
|
41
|
82.8
|
17.2
|
0
|
|
20.7
|
0
|
55.2
|
0
|
0
|
24.1
|
NR
|
100
|
Baldea et al. (2017)31
|
1885
|
23
|
61
|
NR
|
NR
|
NR
|
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
41.7
|
Chaudhry et al. (2017)32
|
23
|
23
|
61
|
0
|
100
|
0
|
|
NR
|
NR
|
NR
|
NR
|
NR
|
4.3
|
87.0
|
NR
|
Morhardt et al. (2018)33
|
46
|
49
|
85
|
84.2
|
NR
|
NR
|
|
14.7
|
7.4
|
38.9
|
0
|
25.3
|
13.7
|
NR
|
NR
|
Prattley et al. (2019)34
|
21
|
49
|
86
|
NR
|
NR
|
NR
|
|
4.8
|
66.7
|
19.0
|
47.6
|
NR
|
14.3
|
42.9
|
76.2
|
Torricelli et al. (2020)35
|
39
|
45
|
74
|
56.4
|
15.4
|
15.4
|
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
NR
|
SCI = spinal cord injury; NR = not recorded; SPC = suprapubic catheter; CIC = clean intermittent catheterization.
|
Population Demographics
The pooled population included 4 829 patients, consisting of 2 686 SCI patients and 2 143 non-SCI patients. The majority of patients were included as part of the study by Baldea et al. (2016) [31], which included 1 885 patients in both the SCI and non-SCI cohorts. Removing this study from analysis yielded 901 patients in the SCI patient group and 258 patients in the non-SCI group. Overall, 65.1% (± 10.8%) of patients were male, and the average patient age was 52.0 (± 7.4) years.
A total of 19 articles reported the etiology of SCI in their patient cohorts [11–13, 15–18, 20, 22–25, 27, 29, 30, 32, 33, 35, 36]. When pooled, the majority of SCI patients (63.0%) had a traumatic etiology for their spinal cord pathology. 24.4% of patients had spina bifida and two studies focused entirely on spina bifida patients [25, 32]. 14.2% of patients were reported have other etiologies of spinal cord pathology, including multiple sclerosis, malignancy, and infection.
The methods of bladder management in SCI patients was reported in 22 studies [10–14, 16–27, 29, 30, 33, 34, 36]. The most common method of bladder management was indwelling catheter (35.7%), with other reported methods including clean intermittent catheterization (18.8%), condom catheter (18.7%), and suprapubic catheter (SPC) (12.5%). 11.6% of patients did not require any form of bladder management and were able to void spontaneously.
19 studies reported including patients with SCI and a history of urinary diversion or bladder augmentation [11–13, 15, 16, 18, 20–27, 30, 32–34, 36]. Of these studies, 16.5% of patients had a history of urinary diversion while 17.6% of patients reported a history of bladder augmentation.
The results of preoperative urine cultures were reported in 16 studies, with 48.1% of patients with SCI demonstrating preoperative bacteriuria [10–12, 14–16, 18, 20, 21, 23–25, 28, 30, 31, 34]. Ten of these papers reported that >90% of their SCI patients had positive preoperative urine cultures [10–12, 14, 16, 20, 23–25, 30].
Kidney Stone Characteristics
The laterality of treated stones was reported in 21 studies, with 21.1% of procedures involving treatment of bilateral kidney stones [10–18, 20–28, 30, 33, 35]. 13 studies described the location of treated kidney stones [11, 12, 14, 16, 18, 21, 22, 24, 26, 27, 29, 33, 34]. 75.5% of procedures were performed on renal stones while 16.2% of procedures were performed on ureteric stones.
Kidney stone size was described in 14 studies [11, 15, 16, 18, 22–24, 27–30, 32–34]. The median stone size was 16.1mm (IQR, 15.6-33.1mm) and 33.8% of procedures involved treatment of staghorn stones. 16 studies reported on kidney stone composition, with 68.1% of analyzed stones being composed of struvite [10, 12–14, 16, 17, 20, 21, 23, 24, 27–29, 33, 34, 36].
Stone-Free Rate
25 studies reported the SFR following surgical therapy for nephrolithiasis for a total of 977 procedures (Supplemental Table S2) [10–26, 28–30, 32–36]. The pooled SFR in SCI patients was 54.1% for patients undergoing SWL, 73.6% for patients undergoing PCNL, and 36.2% for patients undergoing ureteroscopy.
Three studies compared SFR following PCNL in SCI patients to non-SCI patients [14, 28, 35]. Meta-analysis found that there were no significant differences between groups in post-PCNL SFR (OR 0.64; 95% CI, 0.15-2.64; I2 = 77%, p = 0.01) (Fig. 2).
Postoperative Complications
23 studies reported the rate of complications following surgical therapy for nephrolithiasis (Supplemental Table S2) [10, 11, 13–17, 20–35]. Meta-analysis was performed on four studies reporting post-PCNL complications and found that patients with SCI reported a significantly greater rate of grade I (OR 9.54; 95% CI, 3.06-29.79; I2 = 61%, p = 0.08), grade II (OR 3.38; 95% CI, 1.85-6.18; I2 = 58%, p = 0.07), and grades III-V complications (OR 2.38; 95% CI, 1.35-4.19; I2 = 50%, p = 0.11) following PCNL compared to the non-SCI cohort (Fig. 3-5).
When infectious complications were pooled, patients with SCI were more likely experience infectious complications compared to SCI patients (OR 6.15; 95% CI, 1.86-20.39; I2 = 90%, p < 0.00001) (Fig. 6). Types of complications pooled in this analysis included fever, UTI, pneumonia, abscess, wound infection, and sepsis. Specific complications reported in comparative studies are described in Table 2.
Table 2
Summary of the most common postoperative complications reported in comparative studies examining percutaneous nephrolithotomy in patients with and without spinal cord injury
|
|
Complication
|
SCI
|
Non-SCI
|
Grade I
|
Fever
|
43/87 (49.4%)
|
17/226 (7.5%)
|
|
Acute kidney injury
|
5/91 (5.5%)
|
15/161 (9.3%)
|
|
Ileus
|
2/52 (3.8%)
|
0/161 (0%)
|
|
|
|
|
Grade II
|
Bleed requiring transfusion
|
26/126 (20.6%)
|
19/304 (6.3%)
|
|
Urinary tract infection
|
4/39 (10.2%)
|
2/78 (2.6%)
|
|
Pneumonia
|
101/1972 (5.1%)
|
55/2111 (2.6%)
|
|
Wound infections
|
2/52 (3.8%)
|
2/161 (1.2%)
|
|
Venous thromboembolism
|
1/91 (1.1%)
|
1/161 (0.6%)
|
|
|
|
|
Grade III
|
Hemothorax
|
1/126 (0.8%)
|
2/226 (0.9%)
|
|
Abscess
|
3/87 (2.4%)
|
1/226 (0.4%)
|
|
Complication requiring another procedure*
|
16/87 (18.4%)
|
19/304 (6.3%)
|
|
|
|
|
Grade IV
|
Sepsis
|
147/2011 (7.3%)
|
67/2189 (3.1%)
|
|
Acute respiratory distress syndrome
|
3/87 (3.4%)
|
0/226 (0%)
|
|
Myocardial infarction
|
21/1937 (1.1%)
|
22/2045 (1.1%)
|
|
Cerebrovascular accident
|
30/1885 (1.6%)
|
5/1885 (0.3%)
|
|
Multiorgan failure
|
3/52 (5.8%)
|
3/161 (1.9%)
|
|
|
|
|
Grade V
|
Death
|
81/2011 (4.0%)
|
60/2111 (2.8%)
|
SCI = spinal cord injury; * = urinary fistula, pyelocutaneous fistula, nephroenteric fistula, ureteral edema, urinoma/hematoma requiring drainage, nephrostomy tube dislodgement, ureteropelvic junction (UPJ) obstruction, and UPJ injury.
|
Due its disproportionately large study population, sensitivity analysis performed by removing the study by Baldea et al. (2016) [31] from the meta-analysis. Excluding this study did not have a significant impact on SFR, minor complications, or infectious complications, but did have a significant impact on the observed difference in major complication rates.