Introduction

Heterotopic pregnancy (HP) is defined as the simultaneous occurrence of intrauterine and extrauterine pregnancy [1]. Ectopic pregnancy is generally observed at tubes, but it may also occur at a caesarean scar or have a cervical, ovarian or abdominal localisation in rare instances [2]. Although it is rarely seen during natural conception cycles, the incidence has raised to 1:100 due to the increasing use of assisted reproduction techniques in infertile patients [2, 3]. It has been suggested that tubal injury is the most important risk factor for the development of HP. Other potential reasons include tubal surgery, pelvic inflammatory disease, a history of ectopic pregnancy, endometriosis and in vitro fertilisation [2-4]. The most frequent symptoms of HP are abdominal pain and vaginal bleeding, but patients can occasionally present with acute abdomen resulting from a haemoperitoneum [1]. High-resolution transvaginal sonography (TVUS) is the primary diagnostic method, and management may vary depending on the patient’s haemodynamic status, localisation of the ectopic pregnancy and desire of the family for a child [2].

Here, we discuss the presenting complaints, diagnostic and therapeutic approaches and pregnancy outcomes of 6 patients with HP managed in our clinic.

Material and Method

This retrospective study was carried out at the Department of Gynecology and Obstetrics at our university, was planned in accordance with the Second Declaration of Helsinki (revised in 2008) and was approved by the local ethics committee.

Screening the electronic database of the hospital identified records of 113 cases admitted to the clinic with an ectopic pregnancy (ICD codes; O00.0, O00.2, O00.8 and O00.9) and pregnancy (Z33) between January 1, 2010, and June 30, 2013. Of these, 6 patients with a diagnosis of HP who underwent either surgical or medical therapy were included in the study, and 107 patients with ectopic pregnancy alone were excluded. In all the patients, age at treatment, gestational age, presenting complaints, results of pretreatment evaluations, such as sonography, therapeutic approaches employed and pregnancy outcome were reviewed.

TVUS was employed as a preoperative imaging modality. The size of the mass was calculated by measuring the largest 2 diameters in cases in which a gestational sac or an adnexal mass was detected.

The treatment methods employed were classified into 2 groups: surgical and medical. Surgical treatment was further subclassified as follows: radical surgery via laparotomy or laparoscopy and organ-sparing surgery. The medical treatments were injection of either potassium chloride or methotrexate into the gestational sac.

Results

During the 3-year period, 113 patients with ectopic pregnancy were followed in our clinic. Of these, HP was detected in 6 of the patients (5.31%). The mean age of the patients was 30.83±5.41 (24–38) years, and the mean number of pregnancies was 1.67±1.63 (0–4) in patients diagnosed with HP. As predisposing reasons for HP, one patient had a history of pelvic inflammatory disease, and another patient had a history of tubal factor infertility. Pregnancy was achieved by in vitro fertilisation in 2 patients (unexplained infertility and male factor), by recombinant gonadotropin stimulation in one patient. As far as we detected, only last case that having pregnancy in spontaneous cycle has no risk factor. The presenting complaints in 3 cases were abdominal pain and vaginal bleeding and acute abdomen due to a haemoperitoneum in 2 cases. HP was detected by TVUS in one patient who presented to the clinic for routine control.

Table 1 summarises the demographic and clinical characteristics of the patients. Medical therapy was achieved by visualization of foetal heart pulse cessation through either injection of 5 ml (2 mEq/mL) of potassium chloride or 25 mg methotrexate into gestational sac. These interventions in two patients are performed via transabdominal route by using a 20 G spinal needle. For visualization purpose, transabdominal ultrasonography was used. We applied single dose intramuscular injection of 500 mg OH-progesterone prophylactically for patients who had surgery.

Discussion

HP, which has an increasing incidence due to widespread use of assisted reproduction techniques in infertile patients, is an important obstetric problem that can cause maternal and foetal mortality if not diagnosed at an early stage [5, 6]. It has been suggested that a delay in diagnosis can cause maternal mortality of 1% and foetal mortality of 45–65% [2]. It is important to perform a meticulous pelvic sonography evaluation to detect HP even in cases with a positive pregnancy test and an intrauterine gestational sac on sonography after infertility treatment [1].

Diagnosis of a HP can be challenging and a high index of suspicion is mandatory for early diagnosis and appropriate treatment. High-resolution TVUS is the primary diagnostic method in HP [2]. The findings of a complex cyst in the adnexal area, a haematosalpinx, a tubal ring and a foetal heart rate and free fluid in the Douglas space should raise the suspicion of a HP [7].

The serum beta-human chorionic gonadotropin (ß-hCG) level, which is an important laboratory tool for the diagnosis of an ectopic pregnancy, is not considered as a useful marker in HP due to the presence of intrauterine pregnancy. In HP, the ß-hCG level can be detected as normal or elevated [3]. In the present study, all the cases were diagnosed using TVUS. However, the ß-hCG levels were not evaluated. Therefore, the ß-hCG levels cannot be used in the differential diagnosis.

In cases of HP with tubal localisation, patients can present with acute abdomen and hypovolemic shock secondary to tubal rupture [1]. In such cases, the differential diagnosis includes a haemorrhagic corpus luteum cyst, adnexal torsion, a neoplasm and abortion [1, 2]. The use of TVUS and ß-hCG should be beneficial in the differential diagnosis of such clinical entities.

The main issue in the treatment of the HP is to be as minimally invasive as possible to preserve the developing intrauterine pregnancy. The management of HP may vary depending on the patient’s haemodynamic status, localisation of the ectopic pregnancy, skills of the surgeon and preference of the family regarding intrauterine pregnancy [2]. In general, salpingectomy via laparotomy is the choice of treatment in patients who present with acute abdomen due to tubal rupture. Conservative surgery and medical therapy should be employed in cases without haemodynamic instability and rupture, as well in those who want to have children. Laparoscopic salpingostomy is frequently used in conservative surgery, and injection of potassium chloride and methotrexate into the ectopic gestational sac is used in medical therapy [4, 5, 6, 8]. When a review of the literature, it is obvious that there is no unique attitude in medical treatment of HP. Treatment of KCL, placental tissue continues to develop until birth and it can cause acute bleeding. In the treatment of MTX, there are reports on the varying effects of the drug when used at differing stages of fetal development. The most critical exposure time is between six and eight weeks [5, 6]. Whereas, it is suggested that KCL and MTX is a safe and efficacious treatment method in HP [8]. Salpingectomy via laparotomy was warranted in 2 of the cases in the current study who presented with acute abdomen and excessive free fluid in the Douglas space on the TVUS. Conservative therapy was considered inappropriate in these patients. Foetal reduction was achieved by injections of potassium chloride and methotrexate in 2 cases who had an ectopic gestational sac of 18×36 mm and 20×40 mm in size and a foetal heart beat within the gestational sac but no finding of rupture. Laparoscopic salpinostomy was employed as conservative surgery in 2 other cases who had a tubal foetal heart rate.

One study proposed that the rate of intrauterine pregnancy loss increases 2–3 fold in patients who undergo surgical or medical therapy for a HP compared to those with intrauterine pregnancy alone and that two-thirds of these patients have live births [2]. In the current case series, pregnancy resulted in abortion in 3 cases who underwent surgical and medical therapy, and 3 women had live births. When the outcomes of pregnancy were assessed in the context of the surgical and medical therapy, the rates of healthy pregnancy and abortion were identical in both groups (50%).

In conclusion, a careful pelvic examination for HP plays an important role in reducing both maternal and foetal mortality, particularly in cases in which pregnancy is achieved with treatment of infertility such as assisted reproductive techniques and intrauterine insemination, even in cases of detection of intrauterine pregnancy.

Competing interests

The authors declare that they have no competing interests.

References

1. Sun SY, Araujo JE, Elito JJ, Rolo LC, Campanharo FF, Sarmento SG, et al. Diagnosis of heterotopic pregnancy using ultrasound and magnetic resonance ımaging in the first trimester of pregnancy: a case report. Case Rep Radiol 2012;2012:317592. doi: 10.1155/2012/317592.

2. Ercan CM, Keskin U, Coksuer H, Balikçi E, Ergun A. A case of heterotopic pregnancy in a spontaneous cycle. Perinatal Journal 2012;20(2):63-7.

3. Deka D, Bahadur A, Singh A, Malhotra N. Successful management of heterotopic pregnancy after fetal reduction using potassium chloride and methotrexate. J Hum Reprod 2012;5(1):57-60.

4. Schroeppel TJ, Kothari SN. Heterotopic pregnancy: a rare cause of hemoperitoneum and the acute abdomen. Arch Gynecol Obstet 2006;274(3):138-40.

5. Georgiou EX, Domoney C, Savage P, Stafford M. Heterotopic abdominal pregnancy with persistent trophoblastic tissue. Acta Obstet Gynecol Scand 2011;90(5):551-3.

6. Sijanovic S, Vidosavljevic D, Sijanovic I. Methotrexate in local treatment of cervical heterotopic pregnancy with successful perinatal outcome: case report. J Obstet Gynaecol Res 2011;37(9):1241–5.

7. Elito JJ, Montenegro NA, Soares RC, Camano L. Unruptured ectopic pregnancy: diagnosis and treatment. State of art. Rev Bras Ginecol Obstet 2008;30(3):149-59.

8. Litwicka K, Greco E, Prefumo F, Fratelli N, Scarselli F, Ferrero S, Iammarrone E, Frusca T. Successful management of a triplet heterotopic caesarean scar pregnancy after in vitro fertilization-embryo transfer. Fertil Steril 2011;95(1):291-3.

Introduction

Percutaneous nephrolithotomy (PCNL) was first described in 1976 and since then, efforts have been made to make the procedure safer and more effective [1]. Today, PCNL plays an integral role in managing renal stone disease and morbidity and mortality of the disease and also of the treatment itself, has been dramatically reduced. On the other hand, although accepted as minimally invasive, it is an operation which still poses a significant risk of complications. Clavien system has been modified to grade these complications [2, 3] (Table 1).

With numerous previous studies, pre- and peri-operative factors such as stone size and configuration, percutaneous access number and location, entry performed by radiologist or urologist, presence of hydronephrosis have been investigated as predictors of success rates and complications [4-10]. Attempts to identify the associated variables showed variations among the results which has made it difficult to classify the patients so that the stone free rate (SFR) or complications can be predicted. Aiming for a quick, simple and reproducible method for the prediction of the outcomes of PCNL, the ‘Guy’s stone score’ has been proposed by Thomas et al [11]. They have found that the score correlates with stone free rates but not with complications. The grading system mainly takes into consideration the number of stones, stone location and whether the renal anatomy is simple or abnormal. In this scoring system, calyceal diverticulum stones, staghorn stones and any stone in a patient with a spina bifida or spinal injury are the special circumstances that effect the grading of the stone (Table 2). The score is based not just on the stones targeted for treatment in the particular procedure but on all of the stones and abnormal anatomy defines an abnormal renal anatomy, an abnormal collecting system or a patient with an ileal conduit (i.e. cases where the operating surgeon believes access may be difficult). The authors have concluded that the Guy’s stone score can accurately predict the SFR status and have stated that the insignificant correlation with complications may have resulted due to the small patient group of their study in which 100 patients were included.

On the other hand, Smith et al have recently developed a ‘nephrolithometric nomogram’ in order to predict the treatment success in PCNL in which they have used preclinical data and radiological information as the variables of the nomogram (Figure 1). The results of this ‘The clinical research office of the Endourological Society (CROES)’ study showed that stone burden is the best predictor of SFR and case volume, prior stone treatment, presence of staghorn stone, stone location and stone count were the other associated factors [12]. Using these variables a nephrolithometric nomogram was developed which gives an estimated success rate in a range of 30-90% and was found to have a predictive accuracy.

This single institution retrospective study aims to validate the ability of Guy’s stone score and the nephrolithometric nomogram to predict the success of PCNL and to assess the correlation of the scoring systems with the complications encountered.

Material and Method

Medical files of 1646 adult patients to whom a PCNL operation was performed between 8/2002 and 5/2012 were reviewed. Exclusion criteria included non-opaque stones, patients with a stent, nephrostomy tube or indwelling catheter, patients with active UTI, operations in which the percutaneous access was made pre-operatively by a radiologist, contralateral renal or ureteric stone disease, second look PCNL and same session bilateral PCNL operations.

Data recorded from the medical files included: age, gender, stone size (burden), stone location, presence of hydronephrosis, other concomitant disease, previous shock wave lithotripsy or renal surgery history, American Society of Anesthesiologists (ASA) score, subcostal or intercostal entry, number of access tracts made during the operation, dilatation type (Amplatz dilators/balloon dilatation), operation time, blood transfusion, complications, residual stone status, pre-operative and post-operative hemoglobin (Hb), post-operative renal and blood culture results.

The kidney stones were diagnosed by intravenous pyelography and/or computerized tomography (CT). Stones were classified according to the Guy’s stone score (Table 2). For the nephrolitometric nomogram, stone burden was calculated as the surface area calculated according to the EUA guidelines using the two greatest vertical and horizontal dimensional measures seen on a plain x-ray of the kidney-ureter-bladder (KUB) [13]. In patients without a KUB or intravenous pyelography but with a CT scan, the size of stones were calculated by the same way but by the dimensions obtained from the topogram of the non-contrast computed tomography images. Stones filling the renal pelvis and all of the calyces were defined as staghorn calculi and the area of each stone part located in the calyces and the pelvis was calculated separately and then added [13].

Operation time was derived from the anesthesia chart and defined as the time elapsed in minutes from the induction of anesthesia until the insertion of nephrostomy catheter. The patients were grouped into 4 groups (30-60 min, 61-90 min, 91-120 min, and >120 min). The difference of hemoglobin (∆Hb) was calculated by subtracting the hemoglobin value after surgery from the value obtained within a week before surgery. The patients were grouped into 5 groups (0-0.5, 0.6-1, 1.1-1.5, 1.6-2 and >2 gr/dl). Each surgeon’s annual case volume was counted for each year and the previous year value was used when calculating the nephrolithometric nomogram case/volume score. Nephrolithometric nomogram scores exceeding the upper limit of the nomogram (90%) were recorded as 90% for statistical analysis.

The patients were evaluated using a KUB performed on the morning after the procedure and a CT performed 1 month after the operation for the follow-up. Postoperative outcome was evaluated by this post-operative CT and patients lacking it were also excluded from the study. The residual stone status was evaluated in three categories; stone free (SF), clinically insignificant residual fragments (CIRF, residual fragments smaller than 4 mm) and clinically significant residual fragments (CSRF, residual fragments larger than 4 mm). Success rate was defined as the sum of SF and CIRF.

The type of complication encountered was noted and then classified according to the modified Clavien System [1]. For any patient, only the complication of highest grade was analyzed. According to the elective endoscopic operation preparation protocol of our clinic, all patients included in the study had a pre-operative urine culture and were treated pre-operatively according to the antibiotic sensitivity tests. More than 105 organisms per milliliter detected on a urine culture was used to define an infection and the operations were performed only after achieving a negative urine culture. Again, according to the protocol, all patients received peri-operative prophylactic antibiotic treatment before surgery on arrival at the operation theatre with an intravenous administration of 1 g Cephazolin, a first-generation cephalosporin antibiotic and the antibiotic treatment was continued post-operatively with intravenous Ciprofloxacin 200 mg, bid, during the first post-operative day (POD). If post-operative fever >38°C was encountered, urine cultures from the nephrostomy catheter and two blood cultures were obtained and the intravenous antibiotic treatment was continued. The treatment was later changed according to the antibiotic sensitivity of the cultures. Otherwise, in the case of an uneventful post-operative period, this treatment was continued with per os Ciprofloxacin twice a day for 5 days starting from the post-operative day 2. Patients with known drug allergies, patients under age 18 or with renal insufficiency are exceptions for the above-mentioned peri-operative and post-operative antibiotic treatments and for the standardization of the treatment, these patients were not included in the study.

Patients that did not have medical data of the above mentioned study variables were excluded from the study. A total of 1325 patients fulfilled the needed criteria and were included in the study.

Operations: All of the operations were performed in a single center under general anesthesia. After ureteral catheter insertion in the lithotomy position, the patients were tilted to prone position. The pyelocalyceal system was approached with the insertion of an 18-gauge Chiba needle under fluoroscopic guidance. Tract dilation was achieved either via Amplatz or a high pressure balloon dilator NephroMaxTM (Boston scientific) after placement of a safety guide-wire in place. At the end of the dilation, a 30F renal sheath (Microvasive, Natick, Massachusetts) was placed and rigid nephroscopy was performed. A pneumatic lithotripter was used for stone fragmentation. The stone fragments were mechanically extracted and a nephrostomy catheter (14-20F) was placed at the end of the operation. The nephrostomy catheter was withdrawn following an antegrade pyelography which was performed after the gross hematuria cleared.

Statistical analysis: Statistical analysis of data was performed with Statistical Package for Social Sciences (SPSS) 15.0 for Windows and in addition to descriptive statistical methods (mean, standard deviation), Chi-square and Mann-Whitney tests were used for qualitative and numerical data, as appropriate. P values <0.05 were considered to be statistically significant. For the nephrolithometric nomogram validation, ROC curve analysis was performed to estimate the cut of values, sensitivity, specificity and area under the ROC curve (ROC AUC) values for evaluation of success and the complications encountered.

Results

Table 3 shows the demographic and clinical characteristics of the patients and stones. The mean age of the 1325 patients was 46.3±14.7 (range 18-86). Stone burden was the only pre-operative, operative and post-operative variable that had a statistically significant correlation both with success and also with the complications encountered (p<0.05). Guy’s stone score grade 1 and 2 were associated with success and grade 3 and 4 were associated with complications. The associations were statistically significant (p<0.05). Although neither presence of diabetes mellitus nor cardiovascular disease had a statistically significant association between success and complications, ASA score 3 and 4 were seen to be associated with complications (p<0.05). Operation time >120 minutes, presence of a staghorn stone, intra- and post-operative blood transfusion, duration of the nephrostomy catheter, ΔHb more than 2 gr/dl and and positive post-operative urine/blood culture were the other associated variables related to the complications encountered (p<0.05). Presence and the degree of hydronephrosis had no statistical significant correlation with success or complications. The procedural information of the series is summarized in Table 4 and the post-operative data and outcomes are given in Table 5. The total success rate of the patients was 85% in which 71% was stone free and 14% was with CIRF.

ROC curve analysis performed to the nephrolithometric nomogram data showed that with a cut of value of 80.5% for success, the nomogram has a sensitivity of 71% and specificity of 74% (ROC AUC= 80%). Although the nomogram was not designed to give an estimation of the possibility of a complication, the success rates of the nomogram were further analyzed for an association between the complications encountered. ROC curve analysis showed that the association is not strong. With a 78.5% cut of value, it has a 65% sensitivity and 69% specificity (ROC AUC= 67%) to predict a possible complication.

The overall complication rate of this study was 24% and most of the complications encountered were Clavian grade 1 and 2. Clavian 3a, 3b, 4a, 4b and 5 were seen in 8% (3%, 3%, 1% and 1%, respectively). One of the three patients who died (Clavian 5) had a staghorn stone disease and an ASA score 4 (renal failure, diabetes mellitus and hypertension). The operation time was 140 min. and was stone free at the end of the operation. The operation was uneventful and was performed through lower and middle calyx entries. But massive bleeding occurred after the renal sheath was removed and was treated with embolization of the bleeding lower pole accessory artery. The other two patients also had large stones, operation times > 120 min. and Hb loss more than 2 gr/dl. But embolization was not performed and these patients were followed with blood transfusions. In all of the three patients, bleeding stopped but persistent fever was encountered in the post-operative period. They had all four of the systemic inflammatory response syndrome criteria (body temperature lower than 36°C or higher than 38°C, heart rate greater than 90 beats/min, respiratory rate greater than 20 breaths/min or PaCO2 less than 32 mmHg, white blood cell count greater than 12×109/L or less than 4×109/L, or >10% immature forms) but the nephrostomy and blood cultures showed no microorganism growth. Their conditions deteriorated in spite of the administered broad spectrum antibiotics and medical treatment. Eventually multiple organ failure (MOF) developed leading to death. Table 6 shows the complications and the complication rates encountered.

Discussion

Both success rate and complication rate is important for determination of the surgical outcome of PCNL. Success rates can be easily assessed with a generally used < 4 mm cut off point to define CIRF and the sum of CIRF and stone free rates to define success rate. But complication rate determination shows variability among institutions which hampers comparison of outcome data between institutions and also in the same institution within time. Due to the variations among the reported complications which is mainly caused by classification of the complications as minor or major and with the need of a uniform grading system, the ‘Clavien classification system’ has been adopted and modified for the use in urology [3]. Tefekli et al. [14] were the first to report their peri-operative complications of PCNL using this system. They have also redesigned the complications of some of the previously reported large series in which the complications were categorized as minor and major and found that minor complications corresponded to grade 1 and 2, while major complications to grade ≥ 3 [15, 16]. They have concluded that the system is useful for grading the complications but still needs minor modifications especially concerning auxiliary procedures.

Especially Clavian 1 and 2 complications have differences between institutions due to different antibiotic protocols. In an institution where antibiotics are administered post-operatively as a routine treatment, this antibiotic treatment is not accepted as a deviation from the normal pharmacological treatment. But in another institution where no post-operative antibiotics are routinely used, the same antibiotic treatment is graded as Clavian 1. Again, in an institution where radiological interventions are regularly and successfully performed, a complication might be treated by interventional radiology and general anesthesia might not be needed (Clavian 3a). But in another institution, the same complication may have to be treated by surgery under general anesthesia (Clavian 3b).

But even with these and other weak points of the system, it seems to be the best classification system available up to date and was also used in this study. Complications were seen in 24% of the patients and most of them were low grade (smaller than Clavian 3a) and Clavian 3a and higher complications were seen in 8% of the patients. Post-operative antibiotics were used routinely in the patient group of this study which may explain why Clavian 1 complications were encountered less than Clavian 2 complications (7% and 11%, respectively). On the other hand, the retrospective nature of the study may have limited the ability to capture some of the Clavian grade 1 complications.

Most of the variables of the nephrolithometric nomogram are also present in the Guy’s stone score. Stone count is classified as solitary or multiple in both. Stone location and presence of staghorn stone are also similar variables. Guy’s stone score classifies the patients into simple and abnormal anatomy groups. But on the other hand, renal anatomy is not included as a variable in the nephrolithometric nomogram. Instead, previous stone treatment, stone burden and case volume/year are the additional variables that differ from the Guy’s stone score. The nomogram takes into account stone burden and also the stone number (single vs. multiple) while the Guy’s stone score uses solitary, multiple, partial staghorn and staghorn to describe the stone burden. The nephrolithometric nomogram aims to give an estimation of the success rate as a percentage and the Guy’s score aims to grade the complexity so that success can be predicted.

Thomas et al have found that the Guy’s stone score can accurately predict the SFR after PCNL [11]. In their study which they have described the development and validation of the scoring system, they have found that as the grade increases, the success decreases. Grade 1 stones had an 81%, grade 2: 72.4% grade 3: 35% and grade 4: 29% stone free rates. The overall success rate was 62% and complications were seen in 52% of the patients with most of them Clavian grade 1 (30%). No statistically significant correlation was found between the score and the occurring complications which the authors stated that this might have resulted due to the small sample size.

In this study, the overall success rate was 85% and Guy’s stone score 1 and 2 showed a statistically significant correlation with success. Complication rate was 24% and Guy’s stone scores 3 and 4 had statistically significant correlations with the complications. It was seen that as the score increases, the success rates decreases and the possibility of a complication to be encountered increases (Table 7).

In the study of Smith et al. by which they have developed the nephrolithometric nomogram, stone burden was the best predictor of the stone-free rate. Other factors associated with the stone-free rate were case volume, prior stone treatment, staghorn stone, stone location and stone count. They have reported that the nephrolithometric nomogram has a predictive accuracy for success (AUC 0.76).

In this study when a cut of value of 80.5% for success is taken, the predictive accuracy for success has been found to be 80%. The nomogram has a sensitivity of 71% and specificity of 74%. Keeping in mind that the nomogram was not designed to predict complications, when the nomogram was further analysed for an association between the complications encountered, the ROC curve analysis showed that the association was not strong. With a 78.5% cut of value, it has a 65% sensitivity and 69% specificity (ROC AUC= 67%) to predict a possible complication.

In this study, the total complication rate and also the severity of the complications are similar with the complications reported by Tefekli et al. [14] which was also a retrospective study. In their study, Clavian grade 1 complications were also seen less than grade 2 complications (4% vs. 16.3%, respectively). But in a prospective study of de la Rosetta et al, grade 1 complications were reported higher (11.1%) than grade 2 complications (5.3%) which supports the drawback of a retrospective study in means of the limited ability to identify some of the grade 1 complications [2].

Conclusion

A quick, simple and reproducible method which has a good correlation with the SFR and the complication rates of PCNL will improve accuracy of the preoperative counseling of the patient. This study shows that the nephrolithometric nomogram has a good predictive accuracy for success (AUC= 0.80) and the Guys’s stone score has a good correlation both with success and with complications (p< 0.05) and can be used for pre-operative counseling and decision making.

Competing interests

The authors declare that they have no competing interests.

References

1. Fernstrom I and Johansson B. Percutaneous pyelolithotomy. A new extraction technique. Scand J Urol Nephrol 1976;10(3):257-9.

2. de la Rosette JJ, Opondo D, Daels FP, Giusti G, Serrano A, Kandasami SV, et al. Categorisation of complications and validation of the Clavien score for percutaneous nephrolithotomy. Eur Urol 2012;62(2):246-55.

3. Dindo D, Demartines N, and Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2):205-13.

4. Akman T, Binbay M, Sari E, Yuruk E, Tepeler A, Akcay M, et al. Factors affecting bleeding during percutaneous nephrolithotomy: single surgeon experience. J Endourol 2011;25(2):327-33.

5. Michel MS, Trojan L, and Rassweiler JJ. Complications in percutaneous nephrolithotomy. Eur Urol 2007;51(4):899-906.

6. El-Assmy AM, Shokeir AA, Mohsen T, El-Tabey N, El-Nahas AR, Shoma AM, et al. Renal access by urologist or radiologist for percutaneous nephrolithotomy–is it still an issue? J Urol 2007;178(3 Pt 1):916-20.

7. Turna B, Umul M, Demiryoguran S, Altay B, and Nazli O. How do increasing stone surface area and stone configuration affect overall outcome of percutaneous nephrolithotomy? J Endourol 2007;21(1):34-43.

8. Olbert PJ, Hegele A, Schrader AJ, Scherag A, and Hofmann R. Pre- and perioperative predictors of short-term clinical outcomes in patients undergoing percutaneous nephrolitholapaxy. Urol Res 2007;35(5):225-30.

9. El-Nahas AR, Shokeir AA, El-Assmy AM, Mohsen T, Shoma AM, Eraky I, et al. Post-percutaneous nephrolithotomy extensive hemorrhage: a study of risk factors. J Urol 2007;177(2):576-9.

10. Muslumanoglu AY, Tefekli A, Karadag MA, Tok A, Sari E, and Berberoglu Y. Impact of percutaneous access point number and location on complication and success rates in percutaneous nephrolithotomy. Urol Int 2006;77(4):340-6.

11. Thomas K, Smith NC, Hegarty N, and Glass JM. The Guy’s stone score–grading the complexity of percutaneous nephrolithotomy procedures. Urology 2011;78(2):277-81.

12. Smith A, Averch TD, Shahrour K, Opondo D, Daels FP, Labate G, et al. A nephrolithometric nomogram to predict treatment success of percutaneous nephrolithotomy. J Urol 2013;190(1):149-56.

13. Turk C, Knoll T, Petrik A, Sarica K, Straub M, and Seitz C. EAU Guidelines on Urolithiasis. In: members of the European Association of Urology (EAU) Guidelines Office, editors. EAU Guidelines, edition presented at the 25th EAU Annual Congress. Arnhem, The Netherlands: EAU Guidelines Office; 2011.

14. Tefekli A, Ali Karadag M, Tepeler K, Sari E, Berberoglu Y, Baykal M, et al. Classification of percutaneous nephrolithotomy complications using the modified clavien grading system: looking for a standard. Eur Urol 2008;53(1):184-90.

15. Segura JW, Patterson DE, LeRoy AJ, Williams HJ, Jr., Barrett DM, Benson RC, Jr., et al. Percutaneous removal of kidney stones: review of 1,000 cases. J Urol 1985;134(6):1077-81.

16. Lee WJ, Smith AD, Cubelli V, Badlani GH, Lewin B, Vernace F, et al. Complications of percutaneous nephrolithotomy. AJR Am J Roentgenol 1987;148(1):177-80.