238 Current Medical Imaging Reviews, 2009, 5, 238-241
1573-4056/09 $55.00+.00 ©2009 Bentham Science Publishers Ltd.
Calcium Phosphate Stone Morphology: Plain Radiographic Findings and
Interobserver Variabilities
Husnu Tokgoz
1,*, Ozlem Turksoy
2, Abdulkadir Yildiz
1, Ilker Unal
4, Bulent Erol
1, Sema Toparli
5and
Gorkem Mungan
3Karaelmas University School of Medicine, Departments of Urology1, Radiology2 and Biochemistry3, Zonguldak, Turkey; Çukurova University School of Medicine, Department of Biostatistics4, Adana, Turkey; Harran University School of Medicine, Department of Radiology5 , anlıurfa, Turkey
Abstract: In current study, visual appearance of pure calcium phosphate (CaP) urinary stones by its morphology from
plain radiograph (KUB), were investigated. We reviewed patients with urinary stone disease who underwent stone sur-gery. Among cases with crystallographically analyzed stones; 60 patients with pure CaP renal or ureteral stones (greater than 1 cm) were selected. Two endourologists and a uroradiologist familiar with the radiographic patterns of different types of pure stones were invited separately, to evaluate pretreatment KUB. Four different radiographic patterns were identified mainly with the assistance of previous studies. Smooth edged, homogenous stones with denser opacity were in-cluded in Group 1. Stones with irregular edges and stippled border were inin-cluded in Group 2. Relatively uniform stones with radial striations and moderate density were decided to form Group 3. Calculi with the least radiodensity and loosely aggregated with a lacy structure were included in Group 4. Overall, 68.9% of CaP calculi were included in Groups 1 and 2 by all participants. Best concordance between endourologists and the uroradiologist was demonstrated in the evaluation of smaller calculi (< 150 mm2). On KUB, CaP urinary calculi were mostly seen as dense relatively homogenous stones and sometimes with stippled borders. With an increase in pretreatment stone size, interobserver and interdisciplinary variabil-ities increased.
Keywords: Pure, calcium phosphate, morphology, KUB, urinary stone.
INTRODUCTION
CaP urinary calculi account nearly for 5-13% of all uri-nary calculi and accepted as relatively rigid stones [1, 2]. However, CaP stones do not have specific significant fea-tures on kidney-ureter-bladder (KUB) films, and they are the most likely to be misclassified. For both urologists and radi-ologists, it is difficult to identify and predict CaP stones in KUB. In study by Wang et al., only 20% accuracy was achieved for CaP calculi when five experienced surgeons predicted the stone composition based on the morphologic appearance on KUB [3]. Similarly, Ramakumar et al., ob-served that most frequently misclassified stone was CaP, with only 14% being correctly diagnosed [4].
In our study, we investigated the visual appearance of pure calcium phosphate urinary stones by its morphology from plain radiographs. In addition, interobserver variabil-ities (between two endourologists) and interdisciplinary vari-abilities (between endourologists and uroradiologists) were analyzed.
METHODS
A retrospective review was performed of renal and uret-eral calculi treated in our department between the period January 2002 and December 2008. All stones were obtained from surgical procedures (open lithotomy, ureteroscopic
*Address correspondence to this author at the Çukuranbar Mah. 41. Cad. No: 2/ 35 06520, Balgat/Ankara, Turkey; Tel: +90-312-2847285; E-mail: h_tokgoz@hotmail.com
lithotripsy, percutaneous nephrolithotomy). Only pure CaP calculi were included in the study. “Pure calculi” is defined as having more than 80% components to be crystal from spectrophotometric analysis.
Calculi less than 1 cm in greatest diameter and bladder calculi were excluded, as those stones were less distinct and morphological features could not be demonstrated clearly by participants. Cases were also excluded when gas shadows or superimposed bone obscured the morphological appearance of the calculi on KUB films. Stones obtained from spontane-ous passage after fragmentation by SWL were not included in the study as we thought those fragments would not reflect the whole composition of the calculi in the urinary tract of the patient.
The stones were classified into 4 distinct groups accord-ing to their radiological morphology on KUB films. The arti-cles by Ramakumar et al.; Dretler et al. and Bon et al., guided us for defining the groups [4-6]. Group 1 was defined as: mononodular homogeneous calculi with smooth edges, dentate appearance, denser than 12th rib or spinal transverse process (Fig. 1A); Group 2 as: less dense, multinodular stones with stippled border, irregularly shaped but not uni-formly smooth and “motheaten” appearance in the borders of the stone (Fig. 1B); Group 3 as: relatively uniform stones with radial striations, superimposed stipping and moderately dense smooth surface (Fig. 1C); and Group 4 as: calculi with the least radiodensity, loosely aggregated with a lacy struc-ture, and poorly organized (Fig. 1D).
Calcium Phosphate Stone Morphology Current Medical Imaging Reviews, 2009, Vol. 5, No. 4 239
Finally, two experienced endourologists (EU1 and EU2) and a uroradiologist (UR) were invited to examine and ana-lyze the pretreatment KUB film for each patient. All our participants were well educated about the radiographic in-formation and grouping of stones before analyzing the KUB films. They were invited separately in order to avoid influ-ence on each other. They were also blinded to the patient history, laboratory data, treatment modality and outcome.
The concordance between two participants (combination of any two of endourologists and uroradiologist) was meas-ured using Kappa statistic. In order to investigate the effect of stone burden to concordance between participants, stone burden was categorized into three groups as less than 150 mm2, between 150 mm2 and 300 mm2 and greater than 300 mm2. A two-tailed p value of <0.05 was accepted as statisti-cally significant.
RESULTS
Among cases operated at our institution and having crys-tallographic analysis of extracted stones, 60 patients with pure CaP stones that met the aforementioned inclusion crite-ria, were selected and pretreatment KUB films were re-viewed. The location of stones was: renal pelvis [13
(21.7%)]; upper calyx [8 (13.3%)]; middle calyx [10 (16.7%)]; lower calyx [15 (25%)] and ureter [14 (23.3%)].
Mean and median stone burdens were 1184.08 ± 2765.46 and 225 (56 – 12000) mm2, respectively. Of 60 cases, the stones were located in the right kidney in 27 (45%) patients and in the left kidney in 33 (55%) patients.
The mean and median patient’s ages were 48.47 ± 14.2 and 47.5 (14 – 79) years, respectively. Male to female ratio was 17/ 43 (28/ 71%).
In 68.9% of cases, pure CaP stones represented features of Groups 1 and 2. When the decision of each participant was individually checked; 68.3% (for EU1), 70% (for EU2) and 68.4% (for UR) values were reached.
When concordance between participants in the assump-tion of CaP stone morphology, were analyzed; concordance between 2 endourologists was found higher than concor-dance between endourologists and the uroradiologist (Kappa values were 0.72; 0.64 and 0.58 for concordance between EU2-EU1; EU2-UR and EU1-UR, respectively) (p< 0.001). Highest kappa values signifying best concordance between participants were reached when ureteral CaP stones were evaluated (Kappa values were between 0.53 and 0.80 for
Fig. (1). (A) Smooth radiodense CaP calculi (arrowhead) (B) CaP calculi with stippled border, irregularly shaped, not uniformly smooth and
“motheaten” appearance in the borders (arrowhead) (C) Uniform CaP calculi, radial striations with a superimposed stipping, moderately dense with smooth surface (arrowhead) (D) CaP calculi with the least radiodensity, loosely aggregated with a lacy structure and poorly orga-nized (arrowhead).
240 Current Medical Imaging Reviews, 2009, Vol. 5, No. 4 Tokgoz et al.
ureteral calculi). Concordance between participants were found to increase dramatically with smaller stones, espe-cially when the stone burden was <150 mm2.
DISCUSSION
In 1996, Dretler et al., showed that relatively solid, dense smooth stones had more calcium oxalate monohydrate con-tent and were likely to be more difficult to fragment either by SWL, percutaneous nephrolithotomy (PNL) or uretero-scopic lithotripsy [5]. They reported that less dense calculi with reticulated and stippled pattern included predominantly calcium oxalate dihydrate crystals and were easier to frag-ment. Similarly, for renal CaP stones, as the CaP content increased in composition, the stone free rates after PNL de-creased [7]. So, for a CaP stone, someone should suspect similar radiological morphological appearance like a calcium oxalate monohydrate stone. In our study, 38.3% (30-43.3%) of pure CaP stones carried same morphological features of calcium oxalate monohydrate calculi on X-ray (Fig. 1A). Our main goal in current study was to evaluate if calculi composed of CaP crystals had unique plain radiographic properties.
Conventional plain films are widely used in clinical prac-tice and accepted as an easy, non-invasive diagnostic tool especially in the interpretation of calcium containing urinary calculi. When compared with digital images, plain films were found to have similar accuracy in diagnosis of calculi [8]. Most of the time, a urologist or a radiologist is initially exposed to plain films in clinical practice. During this evaluation, clinical and/or radiological experience is critical, and unique radiological features of different types of calculi, including pure CaP stones, need to be identified for objective interpretation as in the case of Dretler and colleagues’ defini-tion of the “burr” formadefini-tion on plain abdominal films in large calcium oxalate calculi [5].
As the radiological evaluation of urinary calculi is sub-jective, 2 urologists mainly dealing with endourology and a radiologist experienced in plain radiographic evaluation were selected in order to observe interdisciplinary and interob-server variabilities. Although, they were invited separately for independent decision making, the concordance was much more than we expected. Based on the findings of the present study we can assume that the stone size/burden mainly af-fects the concordance between all participants as concor-dance was great for ureteral and/or smaller calculi and uret-eral stones are mostly smaller in size when compared with renal stones. Accordingly, identifying and stratifying pure CaP stones solely by radiological features such as stone di-ameter/burden would be improper.
Densitometric evaluation of pure CaP stones with non-contrast computed tomography (NCCT) and comparison of outcomes with that of stones composed of other calcium salts have been studied previously. Patel et al. retrospec-tively evaluated 13 CaP stones with an average composition of 60 to 75% each, and compared hounsfield units (HU) with HU of calcium oxalate, uric acid and cystine stones [9]. They observed that CaP calculi had the highest HU in NCCT. However, the differentiation of calcium oxalate and CaP stones with the help of conventional NCCT remains a more complicated issue and published data have conflicting
re-sults. Although some investigators reported better identifica-tion with novel techniques such as dual source CT, further studies with larger series are mandatory in clinical setting [10]. When compared with simple KUB imaging, higher doses of radiation exposure and increased costs were the major disadvantages related to all kinds of CT applications. Recently, Johnston and coworkers revealed that all stones seen on CT scout were also seen on KUB imaging [11]. Par-ticularly for CaP stones, preoperative morphological appear-ance either on KUB films or NCCT does not differ treatment algorithm. But clinical follow up with KUB imaging should be considered as a safe, cheap and quick way after initial evaluation with NCCT.
Several limitations of the present study should be con-cerned. Although, the whole stone fragments received from the patients were evaluated by spectrophotometry, we were not sure whether the patients brought all of the fragments for analysis. Another limitation is that we included the cases with stones which had more than 80% components to be CaP crystal. So, crystallographic composition other than CaP might disrupt possible unique morphological features of a 100% CaP stone.
In conclusion, on plain abdominal films, pure CaP uri-nary calculi were nearly 70% seen as dense relatively ho-mogenous stones and sometimes with stippled borders. But, no unique, distinct morphological feature specific for a CaP stone was observed (like burr formation). In addition, in-terobserver and interdisciplinary variabilities increase with the increment in pretreatment stone size in CaP calculi.
ABBREVIATIONS
KUB = Kidney-ureter-bladder SWL = Shock wave lithotripsy
PNL = Percutaneous nephrolithotomy HU = Hounsfield unit
NCCT = Non contrast computed tomography
REFERENCES
[1] Hossain RZ, Ogawa Y, Hokama S, Morozumi M, Hatano T. Uro-lithiasis in Okinawa, Japan: A relatively high prevalence of uric acid stones. Int J Urol 2003; 10: 411-5.
[2] Dallera JE, Chandhoke PS. Epidemiology and incidence of stone disease. Urinary Stone Disease. In: Stoller ML, Meng MV, Eds. The practical guide to medical and surgical management. USA: Humana Press Publications 2007; pp. 27-29.
[3] Wang SC, Hsu YS, Chen KK, Chang LS. Correlation between urinary tract pure stone composition and stone morphology on plain abdominal film. J Chin Med Assoc 2004; 67: 235-8.
[4] Ramakumar S, Patterson DE, LeRoy AJ, et al. Prediction of stone composition from plain radiographs: a prospective study. J Endourol 1999; 13: 397-401.
[5] Dretler SP, Polykoff G. Calcium oxalate stone morphology: fine tuning our therapeutic distinctions. J Urol 1996; 155: 828-33. [6] Bon D, Dore B, Irani J, Marroncle M, Aubert J. Radiographic
prognostic criteria for extracorporeal shock-wave lithotripsy: a study of 485 patients. Urology 1996; 48: 556-61.
[7] Kacker R, Meeks JJ, Zhao L, Nadler RB. Decreased stone-free rates after percutaneous nephrolithotomy for high calcium phos-phate composition kidney stones. J Urol 2008; 180: 958-60. [8] Averch TD, O’Sullivan D, Breitenbach C, et al. Digital
radio-graphic imaging transfer: Comparison with plain radiographs. J Endourol 1997; 11: 99-101.
Calcium Phosphate Stone Morphology Current Medical Imaging Reviews, 2009, Vol. 5, No. 4 241 [9] Patel SR, Haleblian G, Zabbo A, Pareek G. Hounsfield units on
computed tomography predict calcium stone subtype composition. Urol Int 2009; 83: 175-80.
[10] Matlaga BR, Kawamoto S, Fishman E. Dual source computed tomography: a novel technique to determine stone composition. Urology 2008; 72: 1164-8.
[11] Johnston R, Lin A, Du J, Mark S. Comparison of kidney-ureter-bladder abdominal radiography and computed tomography scout films for identifying renal calculi. BJU Int 2009; 104: 670-3.
