Natural growth pattern of sporadic renal angiomyolipoma

Introduction

Renal angiomyolipomas (AMLs) are a type of benign renal neoplasm composed of vascular, fat, and smooth muscle components. They are the most frequent benign renal tumor with a prevalence of 0.2%–0.6% (1). There is also a strong female predilection; 80%–90% occur as a clinically isolated entity while the remainder have an association with a phakomatosis such as tuberous sclerosis (2). Renal AMLs are frequently discovered incidentally when the kidneys are imaged for other indications. It is widely accepted that the use of cross-sectional imaging is increasing year on year and the incidence of detection of renal tumors is also increasing (3). As such, the clinical dilemma of how to approach the incidental finding of a small sporadic AML is becoming increasingly relevant. Early studies have recommended a threshold of 4 cm for intervention in asymptomatic AMLs. Other indications for intervention include the presence of symptoms, the presence in a woman of childbearing age, and the suspicion of malignancy (2,4,5). The current European Association of Urology guidelines recommend treatment in well-selected cases including large tumors, tumors in women of childbearing age, and symptomatic tumors. However, the understanding of AMLs is evolving and a large study from 2016 demonstrated that AMLs > 4 cm do not require intervention based on size alone (6). This review did support a policy of active surveillance and notably also included cases of tuberous sclerosis; however, size threshold for treatment remains controversial (2,7,8). For sporadic AMLs there are currently no guidelines on the frequency of imaging studies or which modality should be used. Strategies tend be institution-dependent and an annual follow-up approach has been adopted broadly (2). We hypothesized that the majority of incidentally detected AMLs grow slowly, if at all, that they remain asymptomatic, and therefore do not require intensive follow-up or early intervention, regardless of initial size. The aim of the present study was to evaluate initial size and growth rates of sporadic AMLs.

Material and Methods

Using the National Integrated Medical Imaging System, a keyword search was performed of three separate hospitals’ imaging records between January 2000 and January 2019. Each center is a large tertiary referral center with a busy urology service. The keywords used were “angiomyolipoma” and “AML.” Each result was examined, and the inclusion and exclusion criteria applied. The inclusion criteria used were: lesions ≥1 cm in size; follow-up imaging of at least 36 months, including ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI); lesions under active surveillance that underwent embolization; and lesions that were embolized but had had previous imaging. Exclusion criteria included: presence of tuberous sclerosis; lesions that underwent embolization in the absence of any prior surveillance imaging; and lesions that were embolized on presentation. Lesions were characterized as an AML after at least one CT or MRI.

Patient demographics including age at first scan, gender, and imaging indications were recorded. We also recorded the modality of the first scan, the size of the lesion at first scan, and subsequently the modality of the second scan, the size of the lesion at the second scan, and the duration between scans. The size of the lesion refers to the greatest diameter.

We defined a “grower” as a lesion growing >0.25 cm/year, as did Bhatt et al. in their study (6). A sub-analysis was performed on female patients based on menopausal status. As we did not have access to patients’ charts, we used age < 50 years as a marker of premenopausal status. This method has been used previously in large-scale gynecological studies (9).

Ethical institutional approval was obtained at the start of the study.

Results

A total of 1134 patients were identified with a radiology report that included a reference to renal AML. Filtering to those who had follow-up scans > 3 years apart gave a total of 95 patients. A further 18 patients were excluded for having a diagnosis of tuberous sclerosis. Fourteen patients were excluded as they had presented with symptomatic AMLs that were treated on presentation. In order to measure natural growth rates, we measured all untreated AMLs. Patient and tumor characteristics are shown in Table 1.

OPEN IN VIEWER

Table 1.

Patient and tumor characteristics.

Variable	Category	Value
Gender
Male		4 (7)
Female		53 (93)
Age at first detection (years)		57 (29–78)
Laterality of lesion
Left kidney		18 (50)
Right kidney		18 (50)
Modality of detection
CT		31
MRI		1
US		25
Follow-up (months)		53 (36–225)
Size of lesion at baseline (cm)
1–2		42 (73.7)
2–4		7 (12.3)
>4		8 (14)
Size (cm)		1.3 (1.0–6.0)
Average growth (mm)
Per month		0.00465
Percentage of initial size		3.27%

Values are given as n (%) or median (range).

CT, computed tomography; MRI, magnetic resonance imaging; US, ultrasound.

A total of 63 patients were identified (55 women, 8 men), which satisfied the inclusion and exclusion criteria. Three patients had more than one AML; however, a number of these lesions did not meet inclusion criteria, giving 64 lesions in total for analysis. Each patient had a variety of either CT, MRI, or ultrasound images reviewed. A total of 49 lesions were followed up by the same imaging modality (CT). Fourteen lesions had one CT and were followed up with ultrasound. One lesion had an MRI and was followed up by ultrasound. No lesions demonstrated concerning internal vascularity on either contrast-enhanced CT, MRI, or Doppler ultrasound. The average age at time of first scan was 56.4 years. Of the recorded laterality, 31 lesions were right-sided and 33 left-sided. The average size of AMLs at first scan was 2.08 cm, with nine lesions being ≥4 cm. The average size of AMLs at follow-up scan was 2.16 cm. The average follow-up period was 65.5 months (range = 26–225 months) with a median follow-up of 54 months. The average rate of growth was 0.015 cm per year, depicted graphically in Fig. 1. As the data were not normally distributed, the median was calculated as a more accurate estimate. The median size of lesions at the time of the first scan was 1.4 cm and the median size at the time of follow-up was also 1.4 cm (Table 2).

OPEN IN VIEWER

Fig. 1.

Average growth rate of angiomyolipomas.

OPEN IN VIEWER

Table 2.

Growth rates of lesions.

	First scan (cm)	Second scan (cm)	P
Mean size	2.08	2.16	0.7898 (Student’s t-test)
Median	1.4	1.4	0.9681 (Mann–Whitney U test)
Range	1–10.8	0.7–13

Two patients in our cohort had bilateral lesions which were eligible for inclusion. The first patient, a 36-year-old man, presented with a symptomatic 4.7-cm left-sided lesion that was embolized on presentation. He was found to have a contralateral 1-cm lesion that did not exhibit any growth over a 65-month period. The second patient, a 42-year-old man, presented with a ruptured left-sided 10.7-cm AML, having had a separate right-sided 10.1-cm AML embolized previously. This lesion was treated on presentation. He was also found to have a left-sided 4-cm lesion which was observed. This grew to 4.4 cm over an 83-month period.

Three lesions were defined as “growers,” that is, growing faster than 0.25 cm/year. Therefore, in our cohort, 95.3% of lesions regressed, did not grow, or grew very slowly. Of these two lesions, the first grew from 1 cm to 2 cm over a period of 48 months, as measured on two separate contrast enhanced abdominal CTs. The latter was a female patient aged 46 years at the time of the first scan. The second patient was a female patient, aged 76 years at the time of the first scan with an AML which grew from 4 cm to 4.9 cm over a period of 42 months, as measured on two separate contrast-enhanced abdominal CT scans. The third patient, a 56-year-old male patient, had an incidental finding of a 4.2-cm lesion and was subsequently lost to follow-up. He presented 26 months later feeling lethargic; on CT he was found to have a ruptured AML, the lesion having grown in size to 5.3 cm. This lesion was subsequently embolized.

Ten lesions in our cohort were > 4 cm in size. Of these lesions, two were defined as “growers.” One of these lesions is the patient who ruptured 26 months after the initial scan, described above. This lesion grew at a rate of 0.5 cm/year. The second lesion was in a 76-year-old female patient with a growth rate of 0.26 cm/year. Of the remaining lesions, four regressed and four grew at a rate of < 0.25 cm/year over a mean follow-up period of 97 months.

Two of the lesions in our cohort progressed to require intervention. Of these, one patient presented with a 10.8-cm lesion but declined treatment at the time. This patient presented with a ruptured AML 103 months after initial presentation. The patient presented with flank pain and a drop in hemoglobin. The lesion measured 13 cm and displayed signs of internal hemorrhage. The growth rate of the AML was 0.24 cm/year. The second patient requiring intervention is described above.

None of the patients in our cohort became pregnant during their follow-up; however, 14 of the patients were women of childbearing age (defined as age < 50 years). Of these patients, one had a lesion defined as a “grower.”

Discussion

The management of renal AMLs is becoming an increasingly common conundrum in the world of radiology. With increasing rates of detection, the issue of surveillance and timing of intervention are being encountered more frequently. The approach to renal AMLs has evolved significantly over the past number of decades. One review from 1986 showed that initially in the 1970s > 90% of AML patients underwent nephrectomy (10). It was subsequently shown that 82% of AMLs > 4 cm in size provoked symptoms and these patients were considered for treatment (9). One of the most recent studies in 2016 did not support the threshold of 4 cm as a trigger for intervention (6). Recent guidelines published in the Clinical Kidney Journal in 2017 adopted an annual follow-up approach in asymptomatic patients with AMLs > 3 cm; however, they did advocate for monitoring every two years in asymptomatic cases < 3 cm in size, even suggesting that stability over time should prompt increasing the interval of monitoring in these cases (11).

Four centimeters is often seen as a threshold size, which prompts either treatment with nephrectomy or embolization, or a more aggressive surveillance strategy. Given the small numbers of patients with AML > 4 cm in our cohort, we do not make any specific recommendation regarding these lesions. Bhatt et al. (6) found no difference in growth rate between lesions > 4 cm in size and those < 4 cm in size; however, this was not statistically significant. It is worth noting that in the present study, the two lesions which progressed to rupture were > 4 cm in size. However, 9/10 lesions > 4 cm displayed either none or minimal interval growth, which correlates with Bhatt et al.’s results (6).

None of the lesions < 4 cm in size in the present cohort progressed to require intervention and only one grew significantly. None of these lesions became symptomatic in this time period. Due to the apparent indolent nature of lesions this size, we would recommend relaxing current surveillance strategies in sporadic AMLs < 4 cm. Our study involved a minimum of three years of follow-up and this should be considered when recommending follow-up on incidental sporadic AMLs.

None of the patients in the present cohort became pregnant at any stage throughout their follow-up. It is assumed that AMLs grow more rapidly during pregnancy, based largely on case reports (12). Again, we make no recommendations as to the management of these lesions in pregnant patients. However, 14 patients were women of child-bearing age. Of this patient group, only one had a lesion defined that grew significantly. This suggests that premenopausal status should not be a factor in determining surveillance strategies; however, the present study has too few patients to make any recommendations in this regard.

There was no difference in the median size of lesion in the initial group and the follow-up group, although this was not statistically significant. Our data show that sporadic AMLs demonstrate little to no growth over time. Only 4.7% lesions displayed any significant growth, which correlates with the study by Bhatt et al. (6).

The present study is limited by its retrospective nature and inter-reporter variability. We did not enquire into symptomatology. In addition, in 14 cases, a different modality was used for follow-up (CT or MRI initially, and then ultrasound). Despite this, the present data are consistent with recent studies (6,11,13). In addition, while the present cohort is smaller than that of Chan et al. (13), the median follow-up is much longer at 54 months, compared to 24 months.

In conclusion, the present study confirms the slow-growing nature of sporadic AMLs. With the rising detection rates of AMLs, we would suggest that an active surveillance approach could be adopted and that current guidelines could be relaxed, particularly in lesions that are < 4 cm in size. We would suggest that two follow-up scans at two years apart showing no growth would be sufficient follow-up.