Twenty-year prevalence of diabetes mellitus and hypertension in patients receiving shock-wave lithotripsy for urolithiasis

Authors


Ben H. Chew, Department of Urologic Sciences, Gordon & Leslie Diamond Health Care Centre, Level 6 – 2775 Laurel Street, Vancouver, BC, Canada V5Z 1M9. email: ben.chew@ubc.ca

Abstract

Study Type – Prevalence (retrospective cohort)

Level of Evidence 2b

What’s known on the subject? and What does the study add?

Shockwave lithotripsy is a common and effective treatment method for kidney stones, but has been associated with long-term complications, namely hypertension and diabetes. We compared the prevalence of these two disease in patients treated with lithotripsy to the background provincial population. Our analyses did not find an association between lithotripsy and the development of these diseases.

Shockwave lithotripsy is an effective treatment modality for urolithiasis. The mechanism of stone communition during lithotripsy as well as the acute complications that occur following this treatment have been well described; however, the long-term consequences of this procedure have not been clearly defined. Diabetes and hypertension have been associated with lithotripsy at 19 years follow-up, though this relationship is controversial. This issue is further complicated by the interrelatedness of metabolic dysfunction and stone disease.

Our data show that there is no association between lithotripsy and the development of either hypertension or diabetes. Patients treated for urolithiasis 20 years ago with shockwave lithotripsy were contacted, and their prevalence of diabetes and hypertension in these subjects was compared to the background population of British Columbia. The analysis also considered whether the properties of shockwaves delivered by the original Dornier HM-3 versus a modified Dornier HM-3 differentially affected the risk of our subjects developing these diseases. We did not find that lithotripsy, let alone the type of lithotriptor, was a risk factor for developing hypertension and diabetes. We postulate that the development of renal calculi in our subjects is more indicative of an overall metabolic syndrome where there is increasing evidence that patients with kidney stones get hypertension and diabetes and vice-versa. The development of these diseases is not related to shockwave lithotripsy, but rather to a systemic metabolic dysfunction.

OBJECTIVES

• To compare the prevalence of hypertension and diabetes mellitus (DM) in patients treated with an unmodified HM-3 lithotripter (USWL) and a second-generation modified HM-3 lithotripter (MSWL) 20 years ago at our Centre with that in the provincial population.

• To determine whether the type of lithotripter was differentially associated with the development of these sequelae.

PATIENTS AND METHODS

• Retrospective review of 727 patients at Vancouver General Hospital who underwent shock-wave lithotripsy (SWL) between 1985 and 1989.

• Our study group was compared with Statistics Canada data describing the provincial prevalence of these diseases.

• Multivariate analysis was performed.

RESULTS

• The response rate was 37.3%.

• There was a greater proportion of overweight and obese individuals in the study group compared with the provincial average.

• In univariate analysis, lithotripsy with an unmodified HM-3 (USWL) was associated with a higher rate of DM than the provincial rate, whereas lithotripsy with the modified HM-3 (MSWL) was not.

• Hypertension was more prevalent in all lithotripsy subjects.

• On multivariate analysis the type of lithotripter was not associated with the development of either sequela.

CONCLUSIONS

• No association between lithotripsy and the development of either DM or hypertension in a multivariate analysis

• Metabolic syndrome may have elevated the prevalence of DM and hypertension observed in our subjects on univariate analysis, which is in keeping with the fact that our study population had statistically higher body mass indices than the provincial rate.

• Lithotripsy using the HM-3 was not associated with increased DM or hypertension.

Abbreviations
SWL

shock-wave lithotripsy

BMI

body mass index

DM

diabetes mellitus

MSWL

modified shock-wave lithotripsy

USWL

unmodified shock-wave lithotripsy.

INTRODUCTION

Shock-wave lithotripsy (SWL) is a primary treatment modality for renal calculi up to 20 mm in size [1]. The Dornier HM-1 lithotripter was introduced in 1980 [2]. The use of SWL to treat urolithiasis increased exponentially throughout the 1980s after the introduction and adoption of the HM-3 lithotripter in 1983. Effective fragmentation of renal calculi can be achieved with this device but it is not free from complications.

The most common immediate complications of SWL are haematuria, infection and pain associated with ureteric stone fragments that are difficult to pass [3]. Other complications after SWL can include nausea, vomiting, ecchymosis, abdominal pain and fever [4]. Complications related to locally induced tissue damage have been reported and include renal subcapsular and intraparenchymal haemorrhage, pancreatitis and splenic rupture, as well as other ureteric or gastrointestinal complications [5–8]. In contrast to case reports reporting pancreatitis after lithotripsy, markers of pancreatic damage have not been observed after lithotripsy using a Modulith SLX lithotriptor [9]. Animal and human studies have found that SWL-induced damage is dose dependent and results from microvascular and tubular damage leading to local inflammation and eventual renal fibrosis [10,11]. Interestingly, Handa et al. [6] recently showed that pausing between shock-wave ramping steps induces vasoconstriction during lithotripsy and is protective against haemorrhagic renal injury.

The long-term complications associated with SWL remain controversial. Despite acute renal injury, hypertension and pancreatitis being reported to occur secondary to SWL, these complications are often reversible or transient [5,7,8,12]. Krambeck et al. [13] reported an increased incidence of hypertension and diabetes mellitus (DM) independent of body mass index (BMI) 19 years after SWL with a Dornier HM-3 lithotripter. It is inferred from this study that lithotripsy should be used with caution patients with stones, particularly those with stone-associated co-morbidities such as cardiovascular, renal and pancreatic disease. Conversely, Sato et al. and Makhlouf et al. [14,15] failed to find a similar association.

Currently, approximately 2500 patients receive SWL at the Vancouver General Hospital Stone Centre each year. From 1985 to 1987, patients at our centre received lithotripsy with a Dornier HM-3 lithotripter using a high-intensity, single-shock protocol (unmodified shock-wave lithotripsy: USWL). From 1988 to 1989, our centre relied on a modified Dornier HM-3 (MSWL), which included a low-pressure twin-shock generator and an increased semi-ellipsoid aperture. These modifications reportedly resulted in a reduction in shock-wave intensity from 900 bars to 600 bars at 18 kV (30% reduction) and a reduction in focal zone length from 15 mm to 12 mm in the axial direction [4]. These modifications narrowed the blast path of the original HM-3 and reduced the shock-wave peak pressure. Given the unsettled state of the literature with respect to the long-term complications of lithotripsy, we sought to determine if lithotripsy was associated with the development of DM and hypertension in patients treated at our Stone Center 20 years ago. Additionally, we were interested in determining if USWL or MSWL was differentially associated with the development of DM and hypertension in our patients, given the affects of modifying the original HM-3 lithotripter.

MATERIALS AND METHODS

This study was reviewed and approved by the University of British Columbia Clinical Research Ethics Board (UBC CREB) and the Vancouver Coastal Health Research Institute (VCHRI). The UBC CREB and VCHRI review and oversee research to assure that it meets ethical principles and that it complies with all applicable regulations and standards pertaining to human subject and data protection. This study was conducted in accordance with the VCHRI institutional guidelines, and the regulations and standards of the UBC CREB, which include: the International Conference on Harmonization Good Clinical Practice: Consolidated Guidelines, the Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects, the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, and UBC Policy 89.

Our database identified 1389 patients who underwent lithotripsy at Vancouver General Hospital between 1985 and 1989. We retrospectively reviewed these patient’s charts. A total of 662 patients with kidney stones were initially excluded using an age threshold of 66 years at time of lithotripsy in the 1980s. For the remaining 727 patients, telephone surveys were conducted between June 2007 and August 2008 using a standard form. Eligible patients self-reported their health characteristics. Patients were asked if they had been diagnosed with DM or hypertension in the period after their initial SWL; information was also gathered regarding current height and weight, smoking status, and family history of diabetes. A subject was deemed positive for DM or hypertension if they were diagnosed by a physician, and required medication to manage the disease. We did not make any distinctions between insulin-dependent diabetes and non-insulin-dependent diabetes.

Because it was difficult to identify patients with known stone disease who had never received lithotripsy in our records, the prevalence of hypertension and DM in our study groups was compared with the prevalence of these diseases reported for British Columbia (BC) by Statistics Canada in the 2005 Canadian Community Health Survey. The median age of patients receiving lithotripsy between 1985 and 1989 was 44.9 years. As such, we age-matched the comparator group by restricting the comparison to data from Statistics Canada for adults age-matched to 55–74 years (DM and hypertension) and over 45 years (BMI) from the Canadian Community Health Survey [16–18].

RESULTS

In all, 727 contact letters were mailed to patients. The patients then completed a telephone questionnaire. Our response rate was 37.3% as 455 patients were ineligible for the study. Subjects were ineligible for the following reasons: deceased (43), incorrect address or telephone number (202), unable to reach (114), refused/unable to consent (35). Telephone questionnaires were completed for 271 patients, and eight subjects were excluded because of a diagnosis of DM before the SWL treatment and two women were excluded for uncertainty of DM status.

Of the 261 patients included in the study 168 were men and 93 were women. There were 151 subjects who received USWL and 110 subjects received MSWL (Table 1). The mean age of all subjects was 45.5 ± 10.7 years at the time of the SWL, and 65.4 ± 11.0 years at the time of the survey (Table 1). A total of 79 subjects had a family history of DM and 166 subjects had a history of smoking (Table 1). The mean BMI of all subjects was 27.2 ± 5.3 (Table 1). Approximately 5% of the patients included in the study had a hypertension diagnosis before SWL. None of the 261 study subjects had DM before SWL.

Table 1.  Demographics and risk factors of study responders
RespondersUSWLMSWLAll
  • *

    History is defined as a family history of diabetes (first-degree relative affected) or any history of smoking.

  • Body mass index (BMI) is calculated as mass (kg)/height (m2).

Subjects 151 110261
Men 101 43144
Women 50 67 117
History*   
 Diabetes mellitus 58 44102
 Smoking 70 66136
Age, years (sd)   
 Lithotripsy 46.5 (11.0) 44.1 (10.3) 45.5 (10.7)
 Survey 67.3 (11.1) 62.9 (10.3) 65.4 (11.0)
BMI (sd) 26.9 (5.0) 27.5 (5.8) 27.2 (5.3)

Given that elevated BMI positively correlates with development of urolithiasis and hypertension we assessed the proportion of patients receiving lithotripsy whose weight was normal, overweight, or obese based on the subjects’ reported BMI (Fig. 1). Of the subjects undergoing lithotripsy, 37.4% had a normal BMI and 43.6% of the BC population had a normal BMI (Fig. 1). For lithotripsy subjects, 40.5% were overweight and 22.5% were obese relative to the provincial population, where the rates were 36.5% and 16%, respectively (Fig. 1). The difference in BMI observed for the stone-forming subjects relative to the BC population was statistically significant by chi-squared test (P= 0.002). These data indicate that patients receiving lithotripsy for urolithiasis were more likely to be overweight or obese.

Figure 1.

BMI of study subjects relative to the BC population. Individuals were categorized according to their BMI: normal (BMI < 25.0), overweight (BMI 25.1–29.9), and obese (BMI > 30.0). The relative proportions of people in the normal, overweight, and obese categories were: Lithotripsy, 37.4%, 40.5%, 22.5%; BC population, 43.6%, 36.5%, 16%. A chi-square test (p-value 0.0062) was used to determine if the BMI of lithotripsy patients relative to the BC population was significantly different. Statistically significant differences are indicated (*).

To determine if lithotripsy increased the risk of developing DM and hypertension in our subjects, we performed a univariate analysis to compare the prevalence of these diseases in our subjects relative with the general population. Because gender is differentially associated with metabolic syndrome and cardiovascular disease, we subcategorized our USWL and MSWL subject groups accordingly. The prevalence of DM in men who underwent USWL between 1985 and 1987 was 28.2% whereas 14.9% of men receiving MSWL from 1988 to 1989 developed DM in the 20-year follow-up period (Fig. 2a). Women developed DM at a lower rate. Only 20% and 11.6% of female subjects who underwent USWL or MSWL acquired the condition (Fig. 2a). The rate of DM in the BC population was 10.1% and 9.1% in men and women, respectively (Fig. 2a). This analysis resulted in eight individual comparisons between these groups; as such, we applied a Bonferroni multiple comparison correction to set a significance threshold of P < 0.006. Only the male USWL group was found to have a statistically significant elevation in DM prevalence (P= 0.002). The P value of the female USWL group (P= 0.043) was not significant; nor was there a significant difference in DM prevalence between USWL and MSWL subjects (P= 0.038 for men and P= 0.176 for women) (Fig. 2a).

Figure 2.

Univariate analysis of DM and hypertension prevalence. Patients were grouped according to which lithotriptor they were treated with (USWL or MSWL). The prevalence of (A) DM in males was: USWL, 28.2%; MSWL, 14.9%; Population, 10.1%. In females the prevalence was: USWL, 20.0%; MSWL, 11.6%; BC population, 9.1%. The prevalence of (b) hypertension in males was: USWL, 52.7%; MSWL, 41.9%; BC Population, 30.1%. In females the prevalence was: USWL, 49.0%; MSWL, 50.0%; BC Population, 30.8%. Odds Ratios were calculated using a log-rank test. A bonferroni multiple comparison correction was used to determine significance (p-value 0.00625). Statistically significant difference are indicated (*).

The prevalence of hypertension in male subjects receiving either USWL or MSWL was 52.7% and 41.9%, respectively; for women the difference was negligible (49.0% and 50.0%) whereas the BC population had a prevalence of 30.1% and 30.8% (Fig. 2b). Similar to the prevalence of DM in our subject group, after applying a Bonferroni correction, the only statistically significant difference detected was for men receiving USWL relative to the general population (P= 0.002, odds ratio 2.631, 95% CI 1.472–4.703) despite the prevalence of hypertension in all lithotripsy subjects being generally elevated (Fig. 2b).

We subsequently performed a multivariate analysis using the Cox proportional hazards model to determine the independent risk factors for the development of DM or hypertension in our data set. The type of lithotripsy a subject received, either USWL or MSWL, was not associated with the development of DM (P= 0.648, Hazard Ratio 1.169, 95% CI 0.598–2.284) and hypertension (P= 0.268, Hazard Ratio 0.792, 95% CI 0.525–1.196) (Table 2). Risk factors significantly associated with the development of DM were age (P= 0.031), family history (P= 0.013), and BMI (0.001) (Table 2). While a history of smoking was associated with hypertension in our data set (P= 0.003), age (P= 0.653) and gender (P= 0.439) were not (Table 2).

Table 2.  Multivariate analysis of lithotripsy subjects
FactorHazard ratio95% confidence intervalseP value*
  • *

    P value <0.05 is considered significant. BMI, body mass index; FHx DM, family history of diabetes mellitus; se, standard error; USWL/MSWL, unmodified/modified HM-3 shock-wave lithotripsy.

Diabetes mellitus    
 USWL/MSWL1.1690.598–2.2840.3420.648
 Gender0.5810.286–1.1770.3610.132
 Age1.0351.003–1.0680.0160.031
 FHx DM0.3520.154–0.8020.4200.013
 BMI1.0961.041–1.1550.0270.001
Hypertension    
 Age1.0040.985–1.0240.0100.653
 Gender1.1840.772–1.8170.2180.439
 USWL/MSWL0.7920.525–1.1960.2100.268
 Smoking1.0161.006–1.0260.0050.003

DISCUSSION

Hypertension and DM were first identified as long-term complications of lithotripsy when Krambeck et al. [13] presented evidence that these diseases were associated with renal and proximal ureteral stones treated by lithotripsy at 19 years follow-up [13]. Subsequent to this study, Sato et al. [14] elegantly compared patients with ureteropelvic junction and renal stones relative to patients with distal ureteric calculi and did not find an association between lithotripsy and DM or hypertension. One year later, Makhlouf et al. [15] also failed to find an association between lithotripsy and hypertension or DM; however, only lithotripsy patients treated with an HM-3 lithotripter at 6 years after lithotripsy were used in the study. As a result of these reports, we sought to evaluate whether or not new onset DM and hypertension occurred in relation to lithotripsy at our centre. To do this, we compared our subjects with the BC population because we were unable to identify patients who did not undergo lithotripsy 20 years ago.

Our study is not without limitations. First, there are two limitations to our choice of comparator group that should be discussed. The Health Survey data set does not permit us to isolate factors specific to stone-formers receiving SWL. For example, elevated BMI is independently associated with an increased risk of type II DM as well as with urolithiasis [19,20]. Also, subjects in our study group were likely to be represented in the Health Canada data set and contribute to the province-wide prevalence of hypertension and DM. However, given the limited size of our study group relative to the BC population, we think this is an adequate comparator group. There are additional limitations of this study. It is retrospective in nature, and we had a relatively low responder rate. Also, the clinical diagnosis of DM and hypertension in our subjects might not be consistent with the self-reporting of these illnesses over the telephone. However, it has been shown that self-reporting of cardiovascular disease positively correlates with actual disease presence [21,22]. Lastly, we did not look at the influence of the number of SWL sessions the patients had, to establish if and how more than one treatment correlates with DM and hypertension.

Our univariate analysis of disease prevalence relative to the BC population suggested that MSWL was associated with developing DM; whereas hypertension was associated with USWL and MSWL. However, we were ultimately unable to find that a particular modality of lithotripsy was an independent risk factor for these sequelae on multivariate analysis. Our results are therefore not consistent with Krambeck et al.’s initial findings [13] and coincides with subsequent reports suggesting that these complications do not follow lithotripsy [14,15]. It is important to note; however, that the metabolic syndrome affects many systems concurrently. Urolithiasis, metabolic dysfunction in the form of obesity, endocrine abnormalities associated with stone formation and metabolite disturbances and co-morbid conditions such as hypertension occur as individuals age. Therefore, it is possible that the development of these conditions in our subjects was merely the result of an ongoing multi-factorial metabolic process. This is supported by our BMI data, which show that patients receiving either USWL or MSWL had a significantly elevated BMI relative to the general population.

Furthermore, there is good evidence that there is a strong association between diabetes and the development of kidney stones. Certainly, there is an increased risk of uric acid stones in diabetics compared with non-diabetics. In a study by Pak et al. [23], 33.9% of stones in diabetics were uric acid compared with only 6.2% in non-diabetics. This is consistent with reports finding an association between obesity, diabetes and urolithiasis [20]. Because of the highly linked association between these two diseases, it is difficult to say that there is a causal relationship between shock-wave lithotripsy resulting in diabetes. It is much more likely that obesity and diabetes result in systemic and renal defects, which result in kidney stones, because our patients with kidney stones were more likely to be obese, which is a known risk factor for developing diabetes [20]. Certainly, our data do not show any statistical link of shock-wave lithotripsy to diabetes or hypertension.

Complications of lithotripsy are related to the mechanism by which stones are fragmented. In brief, shock waves are generated by an electrohydraulic, electromagnetic or piezoelectric transducer, transmitted through a liquid medium, coupled to the body, and fluoroscopically targeted to the calculus [1,24–26]. Stone comminution is thought to be achieved by a variety of mechanisms including the generation of tensile forces by impaction of a focused shock wave on the stone leading to spallation, tensile stress with fragmentation, and cavitation [26]. Shock waves damage local vasculature leading to haemorrhage and cavitation in local tissues [24]. Local haemorrhage provides a fluid medium for the generation of cavitation bubbles, which probably damage tissue [6]. Under the right circumstances, lithotripsy might contribute to the development of DM; for example, when the blast path of high-intensity shock waves proceeds directly through the organ, but the pancreas would need to receive a sufficiently damaging shock-wave dose to cause haemorrhage and cavitation. For DM, it has been proposed that glucose intolerance and type 2 diabetes follow from, or contributed to, microvascular changes leading to pancreatic islet ischaemia and beta-cell dysfunction [27]. Disruptions to islet blood flow might then contribute to impaired glucose sensing in glucose-intolerant stone formers in this setting [28]. The current study did not differentiate between type I or type II diabetes and another limitation is that it was not ascertained whether patients were taking oral anti-hyperglycaemic medications or insulin.

We were able to show with univariate analysis that lithotripsy was associated with DM and hypertension; however, this did not hold true in the multivariate analysis when the effect was adjusted for other known risk factors. Our study is based on the first-generation and second-generation Dornier devices, and therefore is not necessarily applicable to the third- and fourth-generation lithotripters currently in use. Lithotripter technology has evolved beyond the Dornier HM-3. There has been a trend to develop lithotripters with higher peak pressures, minimized focal zone size, and improved ultrasonic and fluoroscopic targeting [1,24–26]. It would seem that DM and hypertension are much more likely to occur in people who form stones as a culmination of conventional risk factors unrelated to SWL. As such, rather than dwelling on DM and hypertension as long-term sequelae of SWL, it seems prudent that the focus should be on improving the effectiveness of lithotripsy.

Lithotripsy is not demonstrably associated with the development of DM and hypertension at 20-year follow-up using an unmodified or modified HM-3 lithotripter. Patients receiving SWL for stones had a significantly higher BMI than the provincial population, placing them at higher risk for the metabolic syndrome, which in itself is a risk factor for urolithiasis. Perhaps presentation with a kidney stone is a potential marker of patients that will be clinically diagnosed with DM or hypertension.

CONFLICT OF INTEREST

None declared.