Systematic Review of Caliceal Diverticulum

Abstract

Purpose:

We performed a systematic review of the literature regarding the diagnosis and treatment of caliceal diverticulum (CD).

Methods:

The search strategy was conducted according to the Cochrane review guidelines for systematic reviews and Preferred Reporting Items for Systematic reviews and Meta-Analyses checklist.

Results:

In total 1189 titles and abstracts were reviewed, of which 101 were selected for article evaluation. Of these 101 articles, 40 were selected for inclusion after full article evaluations. In the extracted article, intravenous urography (IVU), rather than CT urography, was the main imaging tool for diagnosis, although many articles had failed to provide any imaging modality information. The extracted articles included 88, 153, and 487 patients who underwent shock wave lithotripsy (SWL), flexible ureteroscopy/retrograde intrarenal surgery (fURS/RIRS), and percutaneous nephrolithotomy (PCNL), respectively. Stone-free rates were 21.3% (SWL), 61.4% (fURS/RIRS), and 83.0% (PCNL). The complication rates were 8.0% (SWL), 3.3% (fURS/RIRS), and 11.9% (PCNL). There was incomplete and inconsistent reporting of even basic clinical parameters, such as the size and location of the CD, number of stones, outcomes, and complications. There was a striking lack of follow-up data, despite a known high recurrence rate. The literature on laparoscopic management was too sparse to analyze.

Conclusions:

This meta-analysis revealed that there are not enough high-quality studies to evaluate the ideal strategy for the diagnosis and treatment of CDs. This systematic review emphasizes (a) the importance of contrast imaging for CD diagnosis, (b) higher success rates but also higher complication rates in PCNL compared with SWL and FURS, and (c) the need for standardized reporting of outcomes to include complications, number of interventions, symptom resolution, stone clearance, and CD ablation.

Introduction

Caliceal diverticulum (CD) definitions and classifications have gone through many alterations throughout the years. The current terminology and definition of CD was made by Prather in 1941.¹ CD has been defined as a congenital, nonsecretory transitional urothelium-lined cavity within the renal parenchyma, usually communicating with the collecting system (calix and rarely renal pelvis) through a narrow diverticular isthmus or “neck”. The theory is that small divisions of the ureteral bud fail to degenerate, resulting in a CD.^2,3 Alternatively, some propose that CD is an acquired condition, secondary to sequelae of obstruction, infections, or fibrosing infundibular stenosis.⁴

CD is a rare disease with an unknown incidence, reported in 0.21%–0.6% of intravenous pyelographic investigations^3
–5 and occurring bilaterally in nearly 3% of patients.⁶ Patients usually present with hematuria, pain, or sepsis, and form stones in up to 50% of cases.^6,7

Stones are thought to be due to urinary stasis, whereby urine fills the cavity, and stagnates due to the narrow neck, precipitating calculi crystals.^8,9 However, an underlying metabolic abnormality for stone formation has been reported in over 50%.^10,11

The critical problem for patients and also physicians is that there has been no consensus on the best treatment strategy to manage CDs. Various treatment modalities have been proposed from conservative treatment, shock wave lithotripsy (SWL), percutaneous nephrolithotomy (PCNL), flexible ureteroscopy/retrograde intrarenal surgery (fURS/RIRS), and partial or total nephrectomy for adults^6,12

–15 and for children.¹⁶ Management tends to follow an upward trend from the less invasive to the more invasive.

We aimed to perform a systematic review of the literature to gain a better understanding of how to diagnose and treat CD, as well as to evaluate the different treatment modality efficacy and safety rates. We also aimed to assess the quality of the published literature to determine whether improvements in reporting and study design are needed.

Methods

Search strategy and study selection

The search strategy was conducted according to the Cochrane review guidelines for systematic reviews and Preferred Reporting Items for Systematic reviews and Meta-Analyses checklist.

Relevant studies were identified from MEDLINE (1966–December 2017), EMBASE (1980–December 2017), Cochrane Central Register of Controlled Trials - CENTRAL (in The Cochrane Library–Issue 1, 2017), CINAHL (1872–December 2017), Google Scholar, and Individual urological journals.

Terms used included: “ureteroscopy,” “flexible ureteroscopy,” “PCNL,” “SWL,” “stones,” “diverticulum,” “urolithiasis,” “calculi,” “calyceal,” and “calyceal diverticulum.”

Mesh phrases included:

(“Diverticulum”[Mesh]) AND (“Calculi”[Mesh])

(“Diverticulum”[Mesh]) AND (“Urinary Calculi”[Mesh])

(“Diverticulum”[Mesh]) AND (“Kidney Calices”[Mesh])

(“Diverticulum”[Mesh]) AND (“Urinary Calculi”[Mesh]) AND (“Kidney Calices”[Mesh])

(“Diverticulum”[Mesh]) AND (“Calculi”[Mesh]) AND (“Kidney Calices”[Mesh]).

Eligibility criteria

All studies reporting their centers' experience with CDs were included. The literature was searched for reviews attempting to classify caliceal diverticula. Only studies on children and adult humans were included. There was no language restriction. References within individual articles were searched for further included studies. Authors of included studies were contacted, wherever data were not available or not clear.

Selection of relevant articles

Two reviewers (O.A. and I.H.) identified studies for potential inclusion. Disagreement between the two extracting authors was resolved by consensus. We limited our inclusion of case series to those reporting on 20 or more to allow for a more robust understanding of the disease from a larger series.

Data extraction and statistical analysis

The first author (I.H.) extracted data from studies relevant to classification of CD, outcomes of various management regimens, and complications. Interventions included SWL, FURS/RIRS, PCNL, laparoscopic surgery, and open surgery. The following variables were extracted: country origin of article, gender, patient's symptom, presence of infection, characteristics of CD and bearing stones, imaging modalities, treatment method outcome: stone-free rates, symptom-free rates, diverticulum-free rates, or no need for further interventions, stone sizes, and complications. To ensure the robustness of the review, the second and third authors (O.A. and F.A.) independently verified all the data.

Similar data were pooled into a cumulative meta-analysis, in an intention to treat basis, to allow a numerical representation of the results.

We aimed to perform a quality assessment of harms using the McHarm scale for each included study.¹⁷ We used Review Manager (RevMan 5.3) to plot the quality assessment of Harms tables.

Results

The literature search flowchart is depicted in Figure 1. In total, 1189 titles and abstracts were reviewed, of which 101 were selected for article evaluation. Of these 101 articles, 40 were selected for inclusion after full article evaluations. Only 1 of 40 studies was an international conference abstract describing a classification system,¹⁸ otherwise all were fully published articles.

FIG. 1.

Flowchart for article selection process of the review.

Overall, 37 of 40 articles reported outcomes for adult CDs, including 24 comparative studies/case reports,^{8

–11,18

–37} 11 review articles,^{1

–6,12

–15,38} and 2 imaging articles.^39,40 The 24 comparative studies/case reports consist of 18 reports of surgical treatment outcome,^{8,19

–35} three reports of classification/treatment algorithm of CD,^18,36,37 and three reports of metabolic analysis of CD patients.^9
–11

Three of 40 articles were for pediatric CDs, including 1 comparative study to report surgical outcomes,⁷ 1 review article,¹⁶ and 1 imaging article.⁴¹

Characteristics of included studies

The included 40 articles were published between 1941 and 2017, with the majority conducted in the USA (26/40), nine conducted in Europe, three in the Middle East, and one each in Australia and Canada. Of 24 comparative studies/case report studies, 18 adult studies^{8,19

–35} and 1 pediatric study²⁸ were conducted to report surgical outcomes and thus enrolled in the meta-analysis in this study. All of these 19 studies were retrospective cohort observational studies, with no randomization or control groups, involving cases between 1983 and 2016. A total of 757 patients (759 renal units) were enrolled and 18 of 19 studies (94.7%) were reports of a single-center experience (Table 1). The methodological quality assessment was difficult due to the disparity of the included studies; however, we aimed to include all the published literature on CD to give a wide picture of the evidence.

Table 1.

The Summary of Enrolled 19 Surgical Comparative or Surgical Case Report Studies (18 Articles for Adult Patients and 1 Article for Pediatric Patients)

Articles	Number of patients	Country of origin	Year of cases	Single/multicenter
Ritchie et al.³⁰	20	United Kingdom	1987–1989	Single
Jones et al.²¹	39 (40 renal units)	USA	1984–1989	Single
Hendrikx et al.³¹	31	The Netherlands	1988–	Single
Batter and Dretler²⁵	26	USA	1989–1995	Single
Bellman et al.²⁰	20	USA	1985–1992	Single
Shalhav et al.²³	30	USA	1987–1996	Single
Donnellan et al.²⁷	21	Australia	1985–1996	Single
Liatsikos et al.⁸	49	USA	1985–1999	Single
Landry et al.²⁴	31	France	1988–1999	Single
Auge et al.²⁶	39	USA	1994–2001	Single
Kim et al.²²	21 (22 renal units)	USA	2001–2003	Single
Turna et al.³²	56	United Kingdom	1989–2004	Single
Estrada et al.²⁸	23	USA	1983–2006	Single
Krambeck and Lingeman⁷	85	USA	2001–2008	Single
Sejiny et al.²⁹	38	France	2003–2009	Single
M Ndez-probst et al.¹⁹	76	Canada	1992–2009	Single
Erkurt et al.³³	47	Turkey	2004–2013	Single
Bas et al.³⁴	54	Turkey	2007–2013	Four referral hospitals
Parkhomenko et al.³⁵	51	USA	2012–2016	Single
Summary	757 (759 renal units)	USA 52.6%, France/Tukey/ United Kingdom 10.5%		Single center 94.7%

There were four studies reporting on the outcomes of SWL, 5 studies on RIRS, 14 studies on PCNL, 1 study on each multitreatment modalities, including PCNL followed by SWL, laparoscopic, and open surgery.

In total, 757 patients were included, of which 60.1% were female (429 women) and 39.9% males (285 men); 20 patients did not have a specific sex reported. The average age reported ranged from 35.4 to 54.7 years of age. There was no predominant side of CD with few bilateral CD (286 right/284 left/1 bilateral/187 unclear). The most frequently present symptom of CD patients was pain (68%–100%). Symptoms also included infections (9%–74%) and hematuria (12.8%–23%). The asymptomatic rate ranged from 0% to 20% (Table 2).

Table 2.

Seven Hundred Fifty Seven Patients' Background Involved in 19 Surgical Comparative/Surgical Case Report Studies (18 Articles for Adult Patients and 1 Article for Pediatric Patients)

Articles	Number of patients	’Patient's age, years (range)	Gender female/male	Side right/left	Infection	Symptom of CD
Ritchie et al.³⁰	20	48 (21–91)	10/10	8/12	Recurrent UTI 20%	Pain 75% No symptom 20%
Jones et al.²¹	39 (40 renal units)	41 (19–66)	23/16	14/26	41%	Pain 71.8% Hematuria 12.8%
Hendrikx et al.³¹	31	—	19/12	—	UTI 38.7%	Pain 80.6% Hematuria 9.7%
Batter and Dretler²⁵	26	39.7	20/6	15/11	Recurrent UTI 54%, Pyelonephritis 12%, Urosepsis 8%	Pain 81%
Bellman et al.²⁰	20	—	—	—	25%	All symptomatic (no details)
Shalhav et al.²³	30	36 (20–58)	19/11	17/13	Recurrent UTI 40%, Febrile UTI 23%	Pain 80% Hematuria 23%
Donnellan et al.²⁷	21	46 (27–74)	9/12	9/12	Recurrent UTI 28.6%	Pain 71.4%
Liatsikos et al.⁸	49	45.6 (9–69)	30/19	27/22	Recurrent UTI 5%, Pyelonephritis 4%	Pain 90%
Landry et al.²⁴	31	41 (19–70)	24/7	16/15	42%	Pain; most frequent
Auge et al.²⁶	39	35.4–35.9	25/14	23/16	Recurrent UTI 28.2%	Pain 100%
Kim et al.²²	21 (22 renal units)	42.4	13/8	14/8	52.4%	Pain 76.2%
Turna et al.³²	56	54.7 (17–82)	31/25	22/34	Chronic UTI 32%	Pain 84%
Estrada et al.²⁸ (For pediatric)	23	5.4 (0.2–12)	14/8	13/10 (bilateral 1)	Febrile UTI 45.5%	Pain 13.6% Hematuria 9.1%
Krambeck and Lingeman⁷	85	42–44	55/30	—	—	—
Sejiny et al.²⁹	38	45.7 (18–72)	22/16	20/18	UTI 23.3%	Pain 100%
M Ndez-probst et al.¹⁹	76	42.5–43.4	49/27	36/40	—	—
Erkurt et al.³³	47	41.8 (16–71)	16/31	26/21	UTI 19.1%	Pain 100% Hematuria 4.3%
Bas et al.³⁴	54	36.3–45.1	28/26	27/27	Recurrent UTI 10.3%–20%	Pain 68%–86.2%
Parkhomenko et al.³⁵	51	44–51	22/7	—	—	—
Total/Range of mean value	757 (759 renal units)	35.4–54.7	429/285 (female/male)	286/284/1 (right/left/bilateral)	Overall infection rate; 9%–74%	Pain 68%–100% Hematuria 12.8%–23% No symptom 0%–20%

— = unclear; CD = caliceal diverticulum; UTI = urinary tract infection.

Imaging

Diagnostic imaging modality for CDs

Table 3 shows the imaging modality used to detect CDs in 19 studies. Intravenous urography (IVU) was reported as the main imaging tool for diagnosis of CDs, although 9 of 19 articles had failed to provide any imaging modality information.

Table 3.

Imaging Modality Used to Detect CDs and Clinical Information About CD and Bearing Stone Involved in 19 Surgical Comparative or Surgical Case Report Studies (18 Articles for Adult Patients and 1 Article for Pediatric Patients)

		Caliceal diverticulum
				Position				Bearing stone
Articles	Imaging modalities	Diameter, mm (range)	Number	U	M	L	Unknown	Diameter, mm (range)	Number	Position
Ritchie et al.³⁰	—	—	—	6	7	7		9 (5–15)	—	Intra CD 100%
Jones et a.²¹	IVU, NCCT, U/S, RP	13 (5–28)	—	15	15	10		—	—	Intra CD 100%
Hendrikx et al.³¹	—	—	—	11	5	15		94 mm²	—	Intra CD 100%
Bellman et al.²⁰	—	—	—	13	6	1		—	—	Intra CD 95% No stone 5%
Batter and Dretler²⁵	IVU, RP	5–17	—	14	5	7		8.2	—	Intra CD 88.5% No stone 11.5%
Shalhav et al.²³	—	—	—	11	15	4		< 15	—	Intra CD 86.7% No stone 13.3%
Donnellan et al.²⁷	IVU, CT, RP	9–90	Single 100%	14	2	5		15 (6–30)	Single 57.1%	Intra CD 100%
Liatsikos et al.⁸	—	22.7 (10–48)	—	32	9	8		17 (2–45)	—	Intra CD 100%
Landry et al.²⁴	—	3–30	—	12	12	7		—	—	Intra CD 100%
Auge et al.²⁶	—	—	Single 97.4%	27	7	6		137–165 mm²	Single 81.1%	Intra CD 94.9% No stone 5.1%
Kim et al.²²	—	15.3 (5–44)	—	12	4	6		138.9 mm²	—	Intra CD 100%
Turna et al.³²	IVU	—	—	26	24	6		11.7–20.6	—	Intra CD 100%
Estrada et al.²⁸ (For pediatric)	IVU, contrast CT	22 (6–70)	Single 100%	17	0	6		—	—	Intra CD 13.6% No stone 86.4%
Krambeck and Lingeman⁷	—	15.8 (5–44)	—	21	18	6	40	—	—	Intra CD 90.6% No stone 9.4%
Sejiny et al.²⁹	IVU, contrast CT	—	—	22	9	7		11.4 (5–22)	Single 63.2%	Intra CD 100%
M Ndez-probst et al.¹⁹	contrast CT, NCCT	—	—	45	16	11	4	583 mm²	—	—
Koopman et al.³⁶	IVU, contrast CT	—	—	60	36	14		—	—	—
Erkurt et al.³³	IVU, contrast CT, NCCT	—	—	31	9	7		21.1 (7–52)	Single 23.5%	Intra CD 100%
Bas et al.³⁴	IVU, NCCT, U/S	—	—	16	15	23		154–211 mm²	—	Intra CD 100%
Parkhomenko et al.³⁵	CT	20.5–28	Single 56.9%	15	11	3		14–16	—	Intra CD 56.9% No stone 43.1%
Total/Range of mean value		13–22.7 mm	Single rate 56.9–100%	360	189	145	44	8.2–21.1 mm (94–583 mm²)	Single rate 23.5%–81.1%	Intra CD 56.9%–100% No stone 5%–43.1%

— = unclear; IVU = intravenous urography; NCCT = noncontrast computed tomography; U = upper pole; M = middle pole; L = lower pole.

Accurate imaging assessment for surgical treatment

Table 3 also showed clinical information about CDs and stone-bearing status in 19 studies. More than half of these articles failed to report CD size and number (Table 2); however, CD position (upper/middle/lower pole) had been described well in almost all of the articles. The location in the upper pole (51.9%) was more common than middle (27.2%) and lower pole (20.9%), as previously reported.^3,4

Surgical treatment options for CDs

Shock wave lithotripsy

Among the studies examined, a total of 88 patients with 89 renal units underwent SWL, of which, the stone-free rate was only 21.3%. Sixty percent of the patients were symptom free after their procedure. The combined complication rate was 8.0% (Table 4). None of the patients had had any treatment before the SWL.

Table 4.

Summary of Treatment Outcome of Shock Wave Lithotripsy for Caliceal Diverticulum

Articles	Number of patients	SFR	Definition of SF	Symptom-free rate	Diverticulum-free rate	No need further treatment rate	Complications	Recurrence	Follow-up	Previous Treatment
Ritchie et al.³⁰	20	40% (8/20)	No residual fragments	75% (12/16)	—	—	Inflammation mass at diverticulum (1/20 5%)	—	3.9 months (0–15.6 months)	—
Jones et al.²¹	15 (16 renal units)	6.25% (1/16)	Clearance of fragments from entire system	56.3% (9/16)	—	—	Hematoma (1/15 6.3xs%), perinephric abscess with nephrectomy (1/15 6.3%)	—	35 months (6–60 months)	—
Hendrikx et al.³¹	15	13% (2/15)	No residual fragments	60% (9/15)	—	60% (9/15)	No major complications 0%	—	—	—
Turna et al.³²	38	21% (8/38) at 3 months after surgery	No residual fragments	61% (23/38) at 3 months after surgery	—	68.4% (26/38)	Febrile UTI (4/38 10.5%)	stone recurrence 1	23.3 months (3–110 months)	—

— = unclear; SFR = stone-free rate.

Flexible ureteroscopy/retrograde intrarenal surgery

A total of 153 patients underwent FURS/RIRS, of which, the stone-free rate was 61.4%, and the symptom-free rate was 67.9%. The complication rate was 3.3% (Table 5). The majority (60.7%) of the patients had had previous failed treatment of other modalities, including SWL and PCNL.

Table 5.

Summary of Treatment Outcome of Flexible Ureteroscopy/Retrograde Intrarenal Surgery for Caliceal Diverticulum

Articles	Number of patients	SFR	Definition of SF	Symptom-free rate	Diverticulum-free rate	No need further treatment rate	Complications	Recurrence	Follow-up	Previous treatment
Batter and Dretler²⁵	26	47.8% (11/23)	No fragment	53.8% (14/26)	—	57.7% (15/26)	Sepsis (1/26 3.8%)	—	45 months (15–84 months)	SWL (18/26 69%)
Auge et al.²⁶	17	19% (3/17)	—	35% (1/17)	18% (3/17)	58.8% (10/17)	Nil 0%	—	6 weeks (5–10 weeks), IVU follow	Nil
Sejiny et al.²⁹	38	55.3% (21/38)	No residual fragment	90% (34/38)	—	—	Significant UTI (3/17 7.9%)	No recurrence	6 months	SWL (30/38 78.9%)
Erkurt et al.³³	47 (55 procedures)	85.1% (40/47) at 3 months after surgery	No residual stone	—	—	—	Febrile UTI (1/47 2.1%)	—	3 months	SWL (32/47 68%), PCNL (4/47 8.6%), Open (2/47 4.3%)
Bas et al.³⁴	25	76% (19/25) at 1–2 months after surgery	No residual stone	92% (23/25) at 3 months after surgery	—	—	Nil 0%	—	3 months	SWL (7/25 28%)

— = unclear.

Percutaneous nephrolithotomy

In total, 487 patients (488 renal units) underwent PCNL as treatment for their CD. Of the extractable data 83.0% (341/411) were stone free, 90.8% (277/305) were symptom free, 11.9% developed complications, and 34.2% (53/161) had had previous treatments (Table 6).

Table 6.

Summary of Treatment Outcome of Percutaneous Nephrolithotomy for Caliceal Diverticulum

Articles	Number of patients	SFR	Definition of SF	Symptom-free rate	Diverticulum-free rate	No need further treatment rate	Complications	Recurrence	Follow-up	Previous treatment
Jones et ak.²¹	14	85.7% (12/14)	Clearance of fragments from entire system	100% (14/14)	100% (14/14)	—	Pleural effusion (1/14 7.1%)	—	35 months (6–60 months)	—
Hendrikx et al.³¹	16	62.5% (10/16)	No residual fragments	62.5% (10/16)	18.8% (3/16)	62.5% (10/16)	Postoperative bleeding (3/16 18.8%), high-grade fever (1/16 6.3%)	Stone recurrence 3	18 months (3–36 months)	—
Bellman et al.²⁰	20	95% (18/19)	—	100% (20/20)	80% (16/20)	100% (20/20)	Hemothorax (1/20 5%), hydrothorax (1/20 5%), persistent extravasations (3/20 15%)	—	6 months (3 months–3 years)	—
Shalhav et al.²³	30	93% (21/23)	—	85% (23/27)	76% (16/21)	—	Hemorrhage (1/30 3.3%), perinephric hematoma and hydrothorax (1/30 3.3%)	Stone recurrence 2, symptom recurrence 4	42 months (12–87 months)	SWL (6/30 20%)
Donnellan et al.²⁷	21	95% (20/21)	—	76.2% (16/21)	30% (6/21)	76.2% (16/21)	Pneumothorax (1/21 4.8%), pyelonephritis (1/21 4.8%), secondary hemorrhage 2 weeks after surgery (1/21 4.8%)	Stone recurrence 3	74.4 months (6–144 months), KUB follow	SWL (9/21 42.9%)
Liatsikos et al.8	49	95.90%	—	100%	87.8% (43/49)	100%	Hemothorax (1/49 2.0%), pseudoaneurysm (1/49 2.0%)	No recurrence	73.2 months (2–147 months)	—
Auge et al.²⁶	22	78% (14/22)	—	86% (19/22)	61% (11/18)	95.5% (21/22)	Significant bleeding (1/22 4.5%), pneumothorax (2/22 9.1)	—	6 weeks (5–10 weeks), IVU follow	fURS (7/22 31.8%)
Landry et al.²⁴	31	84%	—	88%	68% (20/31)	—	Hydro need stents (2/31 6.5%)	Stone recurrence 5	24.6 months (18–96 months), IVU follow	SWL (12/31 38.7%)
Kim et al.²²	21 (22 renal units)	85.7% (18/21)	No residual fragments by noncontrast CT on 1 day after surgery	—	87.5% (14/16) at 3 months after surgery by IVU	—	Nil 0%	—	—	SWL (31.8%), fURS (9.1%), PCNL (9.1%)
Turna et al.³²	18	83% (15/18) at 3 months after surgery	No residual fragments	83% (15/18) at 3 months after surgery	—	94.4% (17/18)	Bleeding during operation necessitating transfusion (2/18 11.1%), pneumothorax (1/18 5.6%)	Stone recurrence 2	48.3 months (12–144 months)	—
Estrada et al.²⁸	6	100% (3/3)	—	100%	—	100%	Nil 0%	No recurrence after surgery	37.2 months (1.2–121.2 months) for treatment Pt	Nil
Krambeck and Lingeman⁷	85	87.8% (65/74)	No residual stone by noncontrast CT on 1 day after surgery	—	73.1% (19/26)	—	Hemothorax (1/85 1.2%), hematoma (2/85 2.4%), acute renal failure (1/85 1.2%), pleural effusion (1/85 1.2%)	No recurrence	—	—
M Ndez-probst et al.¹⁹	76	77%–89%	No stone fragments by KUB+US or noncontrast CT	—	—	—	Pulmonary (21/76 27.6%), hemorrhage (1/76 1.3%), pyrexia (1/76 1.3%)	Unclear	—	—
Bas et al.³⁴	29	82.8% (24/29) at 1–2 months after surgery	No residual stone	93.1% (27/29) at 3 months after surgery	—	—	Bleeding during operation necessitating open conversion (1/25 4%), prolonged urine leakage (2/25 8%)	—	3 months	SWL (8/29 27.6%)
Parkhomenko et al.³⁵	51	96% (49/51) at 1 month after surgery	Clear all stones	100% (51/51) at 1 month after surgery	90.2% (46/51) at 1 month after surgery	—	Hydrothorax (1/51 2.0%), pyelonephritis (1/51 2.0%)	Stone recurrence 1 during 5 year follow-up	35 months	—

— = unclear; PCNL = percutaneous nephrolithotomy.

Others: multiple treatment modalities (PCNL followed by SWL), laparoscopic surgery, and open surgery

Very few studies met our inclusion criteria. Only 1 study for 10 adult patients reported multiple modality treatment, PCNL followed by SWL.²¹ The stone-free rate was 90% and both symptom free and no further treatment rate was 100%. As regards surgical complications, 1 of 10 cases showed pleural effusion (10%) (Table 7).

Table 7.

Summary of Treatment Outcome of Multiple Modality Treatments, Laparoscopic, and Open Surgery for CDs

Articles	Number of patients	SFR	Definition of SF	Symptom-free rate	Diverticulum-free rate	No need further treatment rate	Complications	Recurrence	Follow-up	Previous treatment
Multiple modalities treatment (PCNL followed by SWL)
Jones et al.²¹	10	90% (9/10)	Clearance of fragments from entire system	100% (10/10)	100% (10/10)	—	Pleural effusion (1/10 10%)	—	35 months (6–60 months)	—
Laparoscopic surgery
Estrada et al.²⁸	2	100%	—	100%	—	100%	nil 0%	No recurrence after surgery	37.2 months (1.2–121.2 months) for treatment Pt	Nil
Open surgery
Estrada et al.²⁸	2	100%	—	100%	—	100%	nil 0%	No recurrence after surgery	37.2 months (1.2–121.2 months) for treatment Pt	Nil

— = unclear.

Only one study for four pediatric patients had had laparoscopic/open surgery for their CD, of which all were stone free and symptom free, with no complications²⁸ (Table 7).

Discussions

Out of over 1000 articles identified by our search, only 40 met our inclusion criteria despite being relatively inclusive regarding quality standards. The CD literature is not only sparse, but inconsistently reported, making analysis of outcomes particularly challenging.

Imaging for diagnosis of CDs

Noncontrast-enhanced computed tomography (NCCT) has become the standard for diagnosing acute flank pain or suspected urolithiasis, and has replaced IVU. The advantage of noncontrast imaging must be balanced against loss of information on urinary collecting system anatomy and renal function. Indeed, NCCT is significantly more accurate than IVU in evaluating patients with suspected acute urolithiasis.⁴²

To get an accurate diagnosis of CD, however, a contrast study has to be performed to assess the anatomy of renal collecting system. A previous study reported that CD mostly becomes apparent by IVU because there is inherent connection with the collecting system.⁴⁰ In fact, EAU guidelines only recommend a contrast study as grade A when stone removal is planned and the anatomy of the renal collecting system needs to be assessed.⁴³ Nonetheless, this guideline recommended enhanced computed tomography or IVU just as grade C in complex cases. The widespread adoption of noncontrast CT in place of IVU may delay the accurate diagnosis of CD, and a contrast study should be more strongly recommended, especially in complex cases.

Shortcomings in CD imaging in the literature

Surprisingly, there was a lack of information in the extracted studies of basic anatomical details of CD, size and number of any stones, and imaging modality. Previous treatment algorithms indicated that anterior–posterior position of CD affected the operative outcome and have to be considered in choosing optimal treatment options.^36
–38 These articles recommended laparoscopy for anterior CD and PCNL for posterior CD. However, anterior–posterior information of CD had not been described in the vast majority of articles—only two articles included adequate data.^34,35 We suggest that the surgeon should assess the three-dimensional data of CD for making optimal treatment strategy, including anterior–posterior data, and the future studies of CD should provide the three-dimensional information of CD.

Surgical treatment of CDs

Shortcomings in surgical treatment of CD literature

Apart from one multicenter report,³⁴ the entire literature consists of single institutional experience and no systematic review articles. As a result, the literature is likely to reflect profound institutional bias in patient selection and treatment outcomes. The level of evidence is poor, which is unsurprising given the relatively rare incidence of CD, its wide range of clinical features, and the fact that there are numerous treatment options available. Nonetheless, the quality of reporting in the literature is disappointing due to lack of standardization of CD features, inadequate follow-up, and patchy reporting of complications.

Description of patients and CDs

Existing articles often failed to provide size and location of CD (especially anterior and posterior axis), and size and the number of bearing stone (Tables 4–7). These characteristics of the CD and stones are essential to know which treatment is optimal and to rigorously evaluate these treatment efficacies in the literature. In addition, as mentioned in the imaging section above, the description of CD diagnosis is generally poor in the literature. We believe that reports should include the imaging modality since diagnosis of CD is one of many difficult clinical issues to be resolved.

Treatment outcomes

Most of existing articles reported only a stone-free status as an outcome, ignoring the fate of the CD itself or the patients' symptoms. While most of them determined stone-free status as no residual stone (or no fragment) (Tables 4–7), many articles do not state their definition of stone-free status. Besides, when and how to determine stone-free status is also a crucial point, but again many articles failed to provide these data in the literature. In the case of asymptomatic CD, there is no clear indication for surgical treatment as previous articles suggested.^36
–38 We believe that outcome reporting should include not only stone-free status, but also symptom-free status, and CD-free status. Similar to outcomes after treatment for ureteropelvic junction obstruction, we suggest that any further surgical intervention should be considered a treatment failure.⁴⁴ We propose a categorization of treatment outcomes for CD (Table 8) to comprehensively evaluate treatment outcomes, focused primarily on the need for further surgical intervention, but also including resolution of symptoms and resolution of the CD as demonstrated by imaging follow-up.

Table 8.
Proposed Category of Treatment Outcome for CD

Category Outcome Symptom Caliceal diverticulum Need further treatment

1a Success Nil Nil Nil

1b Success Nil (+) Nil

1c Success (+) Nil Nil

1d Success (+) (+) Nil

2 Failure (+)

Tables 4 –7 showed the follow-up period of each study. Unfortunately, some articles have no data of follow-up period at all; the range of average follow-up periods was 3–74 months. Many studies had inadequate follow-up for a condition with a high recurrence rate, particularly if the CD has not been eliminated. Imaging follow-up for CD should include a contrast study (IVU or contrast CT), however, most of the articles failed to describe follow-up imaging. Thus, the recurrence rate was often not reported in the existing literature to a reasonable standard.

Can we have enough evidence for development of treatment algorithm or clinical classification of CDs?

Anatomical classification system of CD

There were two previous reports proposing anatomical classification systems of CD, one was from Wulfsohn MA,² and the other was from Dretler SP.¹⁸ The former categorized CD into four types by communicating location and shape of the neck, in contrast, the latter classified CD into two types by communicating location, size of CD, and symptom. The categorization of CD in this way, however, is of no practical utility in clinical practice.

Clinical treatment algorithm of CD

Three articles proposed different clinical algorithms which aim to direct the best treatment strategy depending on CD characteristics.^36
–38 These algorithms consist of several different clinical parameters, including symptoms, CD location (anterior/posterior and upper/middle/lower pole), thickness of overlying parenchyma, stone size in CD (cutoff; 15,³⁸ 20,³⁷ and 20–30 mm³⁶). The obvious problem is that there was no evidence base to validate these algorithms and compare their utility.

First, we examined the impact of location of CD on treatment outcome. Some articles showed that the location of CD, such as upper/mid/lower pole, had no influence on treatment outcome of CD in RIRS^26,29 and PCNL.^27,35 However, all algorithms involved location of CD as selection criteria. Furthermore, other stone location information, such as anterior and posterior, was used in the algorithm selection criteria, especially in the selection of laparoscopy (for anterior) and PCNL (for posterior). Nonetheless, there is no clinical data in our study to support this idea. In fact, the majority of articles failed to provide any of these data (see Accurate imaging assessment for surgical treatment section).

Second, we examined the cutoff value of CD size for choosing optimal treatment. Some articles showed there was a significant association between successful outcome and stone size in CD treatment in RIRS,²⁹ although a recent article showed no correlation between success outcome and stone size in CD in PCNL.³⁵ Generally, for renal stone treatment in non-CD cases, the cutoff value of stone size of 20 mm is used to guide treatment selection in EAU guidelines.⁴⁵ However, there is no consensus about legitimate cutoff value of stone size in CD. The cutoff value proposed in these algorithms showed considerable variation (15–30 mm) and is not based on any evidence base, raising doubts about their clinical utility.

Finally, we have to say that there is not enough high-quality clinical data to develop reliable treatment algorithm and validate them for CD treatment. To achieve that, high-quality studies with large number of patients, multi-institutional experience, and prospective study design with standardized reporting are needed.

Conclusions

We aimed to investigate and summarize the current clinical consensus of CD. The most challenging aspects of CD literature are first, the poor quality of diagnostic information, such as size, location, and number of stones, and second the lack of an evidence-based treatment strategy. This meta-analysis revealed that there are not enough high-quality studies to evaluate the ideal strategy for the diagnosis and treatment of CDs. Reasons for the paucity of high-quality studies include the fact that CD is a rare disease and the move away from contrast imaging in stone disease in recent decades. This systematic review emphasizes (a) the importance of contrast imaging for CD diagnosis, (b) higher success rates but also higher complication rates in PCNL compared with SWL and FURS, and (c) the need for standardized reporting of outcomes to include complications, need for and number of further intervention(s), symptom resolution, stone clearance, and CD ablation. Clearly, more high-quality studies are needed before confident assertions about treatment algorithms can be made.

Take home message: Suggested minimum reporting criteria:1. Diagnosis: a. Patient demographics (age, gender) b. Symptoms at presentation c. Imaging modality d. Size and number of stones e. Size and location of CD (anterior/posterior, upper/middle/lower) f. Previous treatment2. Perioperative outcomes: a. Number of procedures b. Length of stay c. Complications using Clavien–Dindo classification3. Postoperative outcomes: a. Length of follow-up b. Need for further intervention c. Stone status d. CD status on imaging e. Symptom status f. Recurrence

Category	Outcome	Symptom	Caliceal diverticulum	Need further treatment
1a	Success	Nil	Nil	Nil
1b	Success	Nil	(+)	Nil
1c	Success	(+)	Nil	Nil
1d	Success	(+)	(+)	Nil
2	Failure			(+)

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Prather

. Calyceal diverticulum. J Urol, 1941; 45:55.

Wulfsohn

. Pyelocaliceal diverticula. J Urol, 1980; 123:1–8.

Middleton

Jr , Pfister

. Stone-containing pyelocaliceal diverticulum: Embryogenic, anatomic, radiologic and clinical characteristics. J Urol, 1974; 111:2–6.

Timmons

Jr , Malek

, Hattery

, et al. Caliceal diverticulum. J Urol, 1975; 114:6–9.

Michel

, Funke

, Tunn

, et al. Pyelocalyceal diverticula. Int Urol Nephrol, 1985; 17:225–230.

Gross

, Herrmann

. Management of stones in calyceal diverticulum. Curr Opin Urol, 2007; 17:136–140.

Krambeck

, Lingeman

. Percutaneous management of caliceal diverticuli. J Endourol, 2009; 23:1723–1729.

Liatsikos

, Bernardo

, Dinlenc

, et al. Caliceal diverticular calculi: Is there a role for metabolic evaluation?. J Urol, 2000; 164:18–20.

Matlaga

, Miller

, Terry

, et al. The pathogenesis of calyceal diverticular calculi. Urol Res, 2007; 35:35–40.

10.

Hsu

, Streem

. Metabolic abnormalities in patients with caliceal diverticular calculi. J Urol, 1998; 160:1640–1642.

11.

Auge

, Maloney

, Mathias

, et al. Metabolic abnormalities associated with calyceal diverticular stones. BJU Int, 2006; 97:1053–1056.

12.

Basiri

, Radfar

, Lashay

. Laparoscopic management of caliceal diverticulum: Our experience, literature review, and pooling analysis. J Endourol, 2013; 27:583–586.

13.

Rapp

, Gerber

. Management of caliceal diverticula. J Endourol, 2004; 18:805–810.

14.

Cohen

, Preminger

. Management of calyceal calculi. Urol Clin North Am, 1997; 24:81–96.

15.

Waingankar

, Hayek

, Smith

, et al. Calyceal diverticula: A comprehensive review. Rev Urol, 2014; 16:29–43.

16.

Silay

, Koh

. Management of the bladder and calyceal diverticulum: Options in the age of minimally invasive surgery. Urol Clin North Am, 2015; 42:77–87.

17.

Santaguida

, Raina

, Ismaila

: McMaster Quality Assessment Scale of Harms (McHarm) for primary studies: Manual for use of the McHarm. Available from: http://hiru.mcmaster.ca/epc/mcharm.pdf

18.

Dretler

. A new useful endourologic classification of calyceal diverticula. J Endourol, 1992; 6(Suppl):81.

19.

M Ndez-Probst

, Fuller

, Nott

et al.: Percutaneous nephrolithotomy of caliceal diverticular calculi: A single center experience. J Endourol, 2011; 25:1741–1745.

20.

Bellman

, Silverstein

, Blickensderfer

, et al. Technique and follow-up of percutaneous management of caliceal diverticula. Urology, 1993; 42:21–25.

21.

Jones

, Lingeman

, Steidle

. The roles of extracorporeal shock wave lithotripsy and percutaneous nephrostolithotomy in the management of pyelocaliceal diverticula. J Urol, 1991; 146:724–727.

22.

Kim

, Kuo

, Tinmouth

, et al. Percutaneous nephrolithotomy for caliceal diverticular calculi: A novel single stage approach. J Urol, 2005; 173:1194–1198.

23.

Shalhav

, Soble

, Nakada

, et al. Long-term outcome of caliceal diverticula following percutaneous endosurgical management. J Urol, 1998; 160:1635.

24.

Landry

, Colombel

, Rouviere

, et al. Long term results of percutaneous treatment of caliceal diverticular calculi. Eur Urol, 2002; 41:474–477.

25.

Batter

, Dretler

. Ureterorenoscopic approach to the symptomatic caliceal diverticulum. J Urol, 1997; 158:709–713.

26.

Auge

, Munver

, Kourambas

, et al. Endoscopic management of symptomatic caliceal diverticula: A retrospective comparison of percutaneous nephrolithotripsy and ureteroscopy. J Endourol, 2002; 16:557–563.

27.

Donnellan

, Harewood

, Webb

. Percutaneous management of caliceal diverticular calculi: Technique and outcome. J Endourol, 1999; 13:83–88.

28.

Estrada

, Datta

, Schneck

, et al. Caliceal diverticula in children: Natural history and management. J Urol, 2009; 181:1306–1311.

29.

Sejiny

, Al-Qahtani

, Elhaous

, et al. Efficacy of flexible ureterorenoscopy with holmium laser in the management of stone-bearing caliceal diverticula. J Endourol, 2010; 24:961–967.

30.

Ritchie

, Parr

, Moussa

, et al. Lithotripsy for calculi in caliceal diverticula?. Br J Urol, 1990; 66:6–8.

31.

Hendrikx

, Bierkens

, Bos

, et al. Treatment of stones in caliceal diverticula: Extracorporeal shock wave lithotripsy versus percutaneous nephrolitholapaxy. Br J Urol, 1992; 70:478.

32.

Turna

, Raza

, Moussa

, et al. Management of calyceal diverticular stones with extracorporeal shock wave lithotripsy and percutaneous nephrolithotomy: Long-term outcome. BJU Int, 2007; 100:151–156.

33.

Erkurt

, Kiremit

, Altay

, et al. Is retrograde flexible nephrolithotripsy feasible for calyceal diverticular stone?. Urolithiasis, 2014; 42:347–351.

34.

Bas

, Ozyuvali

, Aydogmus

, et al. Management of calyceal diverticular calculi: A comparison of percutaneous nephrolithotomy and flexible ureterorenoscopy. Urolithiasis, 2015; 43:155–161.

35.

Parkhomenko

, Tran

, Thai

, et al. Percutaneous Management of Stone Containing Calyceal Diverticula: Associated Factors and Outcomes. J Urol, 2017; 198:864–868.

36.

Koopman

, Fuchs

. Management of stones associated with intrarenal stenosis: Infundibular stenosis and caliceal diverticulum. J Endourol, 2013; 27: 1546–1550.

37.

Miller

, Ng

, Streem

, et al. Laparoscopic management of caliceal diverticular calculi. J Urol, 2002; 167:1248–1252.

38.

Canales

, Monga

. Surgical management of the calyceal diverticulum. Curr Opin Urol, 2003; 13:255–260.

39.

Stunell

, McNeill

, Browne

, et al. The imaging appearances of calyceal diverticula complicated by uroliathasis. Br J Radiol, 2010; 83:888–894.

40.

Rathaus

, Konen

, Werner

, et al. Pyelocalyceal diverticulum: The imaging spectrum with emphasis on the ultrasound features. Br J Radiol, 2001; 74:595–601.

41.

Karmazyn

, Kaefer

, Jennings

, et al. Caliceal diverticulum in pediatric patients: The spectrum of imaging findings. Pediatr Radiol, 2011; 41::1369–1373.

42.

Worster

, Preyra

, Weaver

, et al. The accuracy of noncontrast helical computed tomography versus intravenous pyelography in the diagnosis of suspected acute urolithiasis: A meta-analysis. Ann Emerg Med, 2002; 40:280–286.

43.

Turk

, Petrik

, Sarica

, et al. EAU guidelines on diagnosis and conservative management of urolithiasis. Eur Urol, 2016; 69:468–474.

44.

Vannahme

, Mathur

, Davenport

, et al. The management of secondary pelvi-ureteric junction obstruction—a comparison of pyeloplasty and endopyelotomy. BJU Int, 2014; 113:108–112.

45.

Turk

, Petrik

, Sarica

, et al. EAU Guidelines on Interventional Treatment for Urolithiasis. Eur Urol, 2016; 69:475–482.