Is extracorporeal shock wave lithotripsy the preferred treatment option for elderly patients with urinary stone? A multivariate analysis of the effect of patient age on treatment outcome

Abstract

OBJECTIVES

To investigate the effect of patient age on the stone-free rate (SFR) in patients with urinary calculi treated by extracorporeal shockwave lithotripsy (ESWL).

PATIENTS AND METHODS

In all, 2192 solitary radio-opaque urinary stones of 5–15 mm were identified in adult patients receiving primary ESWL. Patients were divided into three age groups, i.e. ≤ 40, 41–60 and >60 years (579, 1026 and 587 patients, respectively). Multiple logistic regression was used to assess the effect of age and other possible predicting factors (gender, stone characteristics, e.g. side, site and size, and the type of lithotripter used) on the SFR at 3 months after treatment.

RESULTS

The overall adjusted odds ratios (95% confidence interval) for the SFR for those aged 41–60 and >60 years (taking those aged ≤ 40 years as the reference) were 0.708 (0.573–0.875; P = 0.001) and 0.643 (0.506–0.818; P < 0.001). However, if the patients were divided into those with renal or ureteric stones, only the SFR of the former was affected by age, and the adjusted odds ratios were 0.665 (0.512–0.864; P = 0.002) and 0.629 (0.470–0.841; P = 0.002), respectively. Ageing had no effect on the SFR for ureteric stones.

CONCLUSION

The SFR after ESWL for renal stones, but not ureteric stones, was significantly lower in older patients. Further studies on the effects of ageing on renal stone clearance after ESWL are needed to improve stone management in the elderly population.

Abbreviations

FrR

stone fragmentation rate

SFR

stone-free rate.

INTRODUCTION

ESWL was introduced in the early 1980s and despite the initial enthusiasm for ESWL to treat all urinary calculi, the accumulated experience suggested there were some limitations. It was recently reported that elderly patients had a poorer stone fragmentation rate (FrR) after ESWL [1], but the effect of ageing on the outcome of ESWL was not discussed in detail. Therefore, we reviewed possible factors affecting the stone-free rate (SFR) in patients treated by ESWL, with particular attention to the effect of patient age on the outcome.

PATIENTS AND METHODS

Treatment records of adult patients (age ≥18 years) with solitary radio-opaque urinary stones of 5–15 mm and receiving primary ESWL from January 1992 to June 2002 were retrieved from a prospectively collected computer database. The database was started after the establishment of our centre in 1987, and data were collected immediately after ESWL and during the follow-up of the patients, by the staff. During this period patients were treated by one of three lithotripters, the Piezolith 2300 (Richard Wolf, Germany), the MPL 9000 and the Compact Delta (both Dornier MedTech, Germany). Only patients with the stone status available at 3 months after ESWL were included in the study. Patients with a ureteric stent or percutaneous nephrostomy were excluded from the analysis. The patients’ demographic data, stone characteristics, treatment details and treatment outcomes were then analysed. The primary outcome variable assessed was the SFR at 3 months after one session of ESWL, which was defined as the absence of evidence of stone material by plain radiography or ultrasonography at ≤ 3 months after ESWL. In our centre routine radiography immediately after ESWL was used to assess whether there was any evidence of fragmentation or not. The relationship of the success of fragmentation and patient age was also assessed in the analysis.

The treatment protocols for MPL 9000 and Compact Delta were similar; patients treated using these two machines were given oral diclofenac as pre-medication. Additional i.v. fentanyl was used if the patients had pain during treatment. During treatment with these machines the energy level was gradually increased according to the tolerance of the patient. For the Piezolith 2300, because the treatment was less painful, no routine pretreatment analgesia was given and the maximum power level (Level 4) was used for all patients. Only very occasionally was diclofenac given to the patient if they had pain during treatment. Treatment was aimed at a maximum of 4000 shocks for the Piezolith 2300 and Compact Delta, and 1500 for the MPL 9000, unless the stone became difficult to visualize or patients could not tolerate further treatment.

Simple and multiple logistic regression was used to assess all the potential predictive factors of the SFR at 3 months, and the success of fragmentation. Potential predictive factors included patient age and gender, stone characteristics (side, size and site) and type of machine used. Patients were assigned to one of three groups by age, i.e. ≤ 40, 41–60 and >60 years old, with the first group used as the reference category. Stone size was entered as a continuous variable. When evaluating the effects of stone location by multiple logistic regression, the locations were chosen as reference categories if they were close to the mean outcome during univariate analysis. For the effects of the machine used, indicator variables were created for the Piezolith 2300 and MPL 9000 (with the Compact Delta used as the reference category). All covariates were entered into the multivariate model regardless of their significance level in the univariate model. Backward stepwise selection was used for the selection of the final multivariate logistic regression model.

RESULTS

In all, 2489 patients with solitary radio-opaque urinary stones of 5–15mm received primary ESWL from January 1992 to June 2002. Follow-up information of 297 (11.9%) patients was incomplete and therefore only 2192 patients were available for analysis. There were 1498 (68.3%) renal stones and 694 (31.7%) ureteric stones. The mean (range) age of the patients was 50.6 (18–97) years; 579 (26.4%) were aged ≤ 40, 1026 (46.8%) 41–60 and 587 (26.8%) >60 years. The characteristics of patients in each age group are listed in Table 1. The stone distribution between different age groups was comparable, but there was gradual increase in the mean stone size across the age groups.

Table 1.  The characteristic of patients and stones, the treatment characteristics and follow-up methods, and the SFR at 3 months after ESWL and the stone FrR for the overall population, and for renal and ureteric stones in each age group

Variable, n (%)	Age, years			Total
	≤40	41–60	>60
KUB, plain abdominal film; US, ultrasonography.
Number of patients	579 (26.4)	1026 (46.8)	587 (26.8)	2192 (100)
Male	375 (64.8)	781 (76.1)	420 (71.6)	1576 (71.9)
Female	204 (35.2)	245 (23.9)	167 (28.4)	616 (28.1)
Stone site:
Kidney
upper calyceal	36 (6.2)	73 (7.1)	43 (7.3)	152 (6.9)
mid-calyceal	64 (11.1)	93 (9.1)	80 (13.6)	237 (10.8)
lower calyceal	191 (33.0)	381 (37.1)	205 (34.9)	777 (35.4)
renal pelvic	76 (13.1)	155 (15.1)	101 (17.2)	332 (15.1)
Ureter
upper ureteric	149 (25.7)	213 (20.8)	100 (17.0)	462 (21.1)
mid-ureteric	7 (1.2)	14 (1.4)	9 (1.5)	30 (1.4)
lower ureteric	56 (9.7)	97 (9.5)	49 (8.3)	202 (9.2)
Right	334 (57.7)	597 (58.2)	342 (58.3)	1273 (58.1)
Left	245 (42.3)	429 (41.8)	245 (41.7)	919 (41.9)
Stone size, mm
5–10	461 (79.6)	771 (75.1)	400 (68.1)	1632 (74.5)
11–15	118 (20.4)	255 (24.9)	187 (31.9)	560 (25.5)
Treatment and follow-up methods:
Piezolith 2300	255 (44.0)	521 (50.8)	297 (50.6)
MPL 9000	147 (25.4)	215 (21.0)	135 (23.0)
Compact Delta	177 (30.6)	290 (28.3)	155 (26.4)
Mean (median) no. of shocks
Piezolith 2300	3063/3000	3219/3000	3315/3500
MPL 9000	1394/1206	1456/1300	1445/1300
(Mean kV)	19.66	20.08	19.94
Compact Delta	3236/3500	3339/3500	3163/3400
N (%) using power setting ≤ 3	71 (40.3)	99 (34.3)	54 (35.5)
N (%) using power >3	105 (59.7)	190 (65.7)	98 (64.5)
Follow-up
Lost, n (%) of total	87 (13.1)	144 (12.3)	66 (10.0)
Outcome assessed by:
KUB	576	1019	580
US alone	2	6	3
KUB and US	1	1	4
SFR at 3 months and the FrR, n (%) of total and SFR/FrR, %
Renal	367 (63.4)	702 (68.4)	429 (73.1)	1498 (68.3)
	53.4/93.7	40.9/86.8	40.6/82.3	43.9/87.2
Ureteric	212 (36.6)	324 (31.6)	158 (26.9)	694 (31.7)
	55.2/75.9	47.5/73.8	41.1/68.4	48.4/73.2
Total	579	1026	587	2192
	54.0/84.3	43.0/82.7	37.6/76.0	44.5/81.3

The treatment variables for each age group are also listed in Table 1. For the three age groups, the proportion of patients treated by each machine was not exactly the same. While the proportion of patients treated with the MPL 9000 were similar in different groups, there were more elderly patients treated with the Piezolith 2300 and more young patients treated with the Compact Delta. For each machine, the number of shocks received and energy level used in different age groups are also listed. The variables for Piezolith 2300 and MPL 9000 were similar in the three age groups. For the Compact Delta, a higher energy level was used in the older patients. The proportion of patients in each group that did not have complete follow-up information was 10–13.1%. Plain radiography was used in most patients to assess treatment outcome.

The SFR and FrR for the overall population and each age group are also listed in Table 1. The overall SFR at 3 months was 44.5%. The SFR of the three age groups were 54.0%, 43.0% and 37.6%, respectively. Both the SFR and the FrR were lower in older patients.

As the stone characteristics of each age group were not exactly the same, a multivariate analysis was used to assess the effect of individual predictive factors on the outcome. The overall adjusted odds ratios for SFR at 3 months are given in Table 2. Younger patients, smaller stones, left-sided stones and patients treated with the MPL 9000 or Piezolith 2300 were factors associated with a significantly better SFR in the multivariate analysis. Patients with lower calyceal stones had a significantly lower SFR. From these results, patient age was a significant factor for SFR and those aged >60 years had the worst results. However, if the patients were divided into those with renal and ureteric stones, only the SFR for renal stones was affected by age, and not that for ureteric stones. The adjusted odd ratios (95% CI) for renal stones in patients aged 41–60 and >60 years (with reference to those aged ≤ 40 years) were 0.665 (0.512–0.864; P = 0.002) and 0.629 (0.470–0.841; P = 0.002), respectively.

Table 2.  The adjusted odds ratio for predictive variables of SFR at 3 months after ESWL (only significant variables shown)

Variable	Adjusted odds ratio (95% CI)	P
Age, years (reference, ≤ 40 years)
41–60	0.708 (0.573–0.875)	0.001
>60	0.643 (0.506–0.818)	<0.001
Stone side (reference, left)
Right	0.811 (0.680–0.967)	0.020
Stone size	0.860 (0.831–0.890)	<0.001
Machine used (reference, Compact Delta)
Piezolith 2300	1.403 (1.136–1.733)	0.002
MPL 9000	1.751 (1.258–2.257)	<0.001
Stone site, lower calyceal	0.739 (0.577–0.944)	0.016

For the FrR after ESWL, the proportion of patients with evidence of fragmentation in the overall population, renal and ureteric groups were 81.3%, 87.2% and 73.2%, respectively. The FrR of overall, renal and ureteric stone according to age group are listed in Table 1, and the effect of age on these groups is shown in Table 3. In the overall assessment, patients aged >60 years had significantly poorer FrR than the other age groups. In the subgroup analysis for renal and ureteric stones there was no significant difference among different age groups, although there was a tendency for a lower FrR of renal stones in those aged >60 years.

Table 3.  The adjusted odds ratio for the FrR immediately after ESWL for different age groups (with ≤ 40 years as the reference group)

Adjusted odds ratio (95% CI), P	Age, years
	41–60	>60
Overall	0.991 (0.743–1.322), 0.951	0.727 (0.530–0.999), 0.049
Renal stone	0.984 (0.655–1.477), 0.936	0.716 (0.465–1.103), 0.130
Ureteric stone	1.049 (0.697–1.579), 0.819	0.774 (0.485–1.236), 0.284

DISCUSSION

The present results show that patient age was a significant predicting factor for SFR after ESWL of renal stones, and the older the patient the lower the SFR. Besides patient age, a larger stone, lower calyceal stones and right-sided stones were all pretreatment factors that were associated with a significantly lower SFR.

Although there are many reports discussing various factors affecting the clinical outcome of ESWL, patient age is mentioned infrequently. In an early report assessing the prognostic factors for treatment outcome of the HM-3 (Dornier MedTech) for renal stones, patients aged >60 years had the poorest SFR amongst all age groups [2]. In a recent study, 3023 patients with urinary calculi (both renal and ureteric) were divided into four age groups and the results showed that older patients were associated with a significantly poorer SFR [3]. In the report of Abdel-Khalek et al.[4], 2954 patients with renal stones (no ureteric stones) were treated by ESWL. By multivariate analysis, patients aged >40 years had a significantly poorer SFR. However, in another report on ureteric stones by the same group [5], patient age was not a significant factor for the treatment outcome, by analysis using either an artificial neural network or logistic regression model. Similarly, Delakas et al.[6] analysed the treatment outcomes of 688 patients with ureteric stones, and patient age was again not a significant predictor for treatment success. From the present results, patient age was a significant predictor of outcome in the overall analysis and by renal stone subgroup, but not for ureteric stones. Our findings are in keeping with the previous reports; ageing only affects the treatment outcome of renal but not ureteric stones. What is the reason for this observed difference of ageing on ESWL?

Ikegaya et al.[1] reported that the FrR was poorer in the elderly population. In our overall analysis of the FrR immediately after ESWL, patients aged >60 years also had a poorer FrR. Although in the subgroup analysis there was no significant difference among age groups, there was a tendency to a lower FrR of renal stones in those patients aged >60 years. This suggests that ESWL of renal stones in elderly patients might have a poorer FrR, which in turn resulted in poorer treatment outcomes.

The reason for the possible poorer FrR of renal calculi in the elderly population is unknown, but it is unlikely to be related to any change in the composition of the stones, otherwise both renal and ureteric stones would be affected. The other possibility might be related to the effectiveness of the transmission of shock-wave energy to the targeted stone. The main difference in the shock-wave path for renal stones, compared with ureteric stones, is that the shock wave needs to pass through the kidney parenchyma before reaching the stone. It is well known that ageing results in sclerotic changes in the kidneys [7,8]. Sclerotic kidneys could then affect the acoustic impedance of kidneys, as reflected by changes in echogenicity of kidneys with ultrasonography. We suspected that this change in acoustic impedance of ageing kidneys might be one of the reasons for the lower efficacy of shock-wave transmission and hence lower FrR and SFR of renal stones. However, further studies will be needed to confirm this suggestion.

Another cause that might account for the lower SFR in the elderly population is the possibility of changes in pelvi-ureteric motility secondary to ageing. There is no reported evidence suggesting that ageing has an effect on pelvi-ureteric motility. From the experiences of IVU and dynamic radioisotope studies there seems to be no clinically significant difference in the drainage of contrast medium or isotopes among different age groups, but this might be an area worth further evaluation.

The present study can be criticised for having insufficient power to show the effect of patient age on the treatment outcome of ureteric stones. However, our current results seem to be in keeping with those reported previously, and at least the effect of age on the success of renal stone treatment appears to be genuine.

Another possible criticism is the use of plain radiography to assess treatment outcome; this (including tomography) would inevitably result in over-estimation of SFR when compared with unenhanced CT, although such CT was not a standard practice in our centre during the study period (1992–2002) and we had to rely on plain radiography to assess treatment outcome. As most of the patients were assessed by the same radiographic method, this over-estimation of the SFR would probably not affect the interpretation of results.

As ESWL in the elderly patient has a lower success rates, more treatment sessions might be offered to overcome the problem. However, ESWL in the elderly population is not without complications. Janetschek et al.[9] reported that there was more new-onset hypertension after ESWL in patients aged >65 years. Bhatta et al.[10] also reported that elderly patients had a greater chance of subscapsular or perinephric haematoma after ESWL by electromagnetic lithotripters, which is the most commonly used generator design in modern lithotripters. However, in clinical practice, ESWL is frequently used as the first-line treatment for the elderly, to avoid the invasiveness and potential complications of endoscopic treatment and anaesthesia. With the improvement in technique and instruments, the risks of modern endoscopic stone procedures are minimal. For ureteroscopy, the use of small-calibre endoscopes and holmium laser lithotripsy has decreased clinically significant complication rates to <1%[11]. Although the need for regional or general anaesthesia is one of the criticisms of using ureteroscopy rather than ESWL, there are reports of the successful use of local anaesthesia and i.v. sedation for ureteroscopy [12]. There is also a report showing that the complication rate of percutaneous nephrolithotomy in the elderly (age >70 years) was similar to that in the younger population [13]. Therefore, endoscopic treatment should not be withheld in the elderly, especially when the stones are likely to have a poorer outcome from ESWL, e.g. in the lower calyx or larger stones.

In conclusion, patient age was a significant predicting factor affecting the treatment outcome of ESWL for renal, but not ureteric, stones. Further studies are needed to identify possible explanations for this observation. Because of the lower success rate and potential short- and long-term complications of ESWL in the elderly, other treatments should be more readily considered for treating urinary calculi in these patients.

ACKNOWLEDGEMENTS

The authors thank the radiographers, nursing staff, Mr G. Smith, and Dr S. Moussa in the Scottish Lithotripter centre for their effort in patient management, and their hard work in keeping the treatment data for the database.

CONFLICT OF INTEREST

None declared.