Skip to main content
NCBI home page
RSNA Journals logoLink to RSNA Journals
. 2010 Apr 8;255(2):495–500. doi: 10.1148/radiol.10091769

Visualization of Renal Medullary Hyperattenuation at Unenhanced CT: What Is the Effect of Furosemide Administration?1

Rahi Kumar, Zhen J Wang ✉, Yanjun Fu, Carlos Forsythe, Emily M Webb, Benjamin M Yeh
PMCID: PMC4128916  NIHMSID: NIHMS611482  PMID: 20413762

Abstract

Purpose

To retrospectively investigate the effects of furosemide on the visualization of renal medullary hyperattenuation at unenhanced computed tomography (CT).

Materials and Methods

This retrospective single-institution study was HIPAA compliant and approved by the institutional review board; requirement for informed consent was waived. This study identified 289 consecutive patients (152 men, 137 women; mean age, 59 years) without ureteral obstruction who underwent unenhanced scanning as part of CT urography; of these, 178 patients did not receive intravenous furosemide prior to imaging and 111 did. The presence of renal medullary hyperattenuation, renal stones, and bladder urine attenuation levels were recorded and compared between patients who did not receive furosemide prior to imaging and those who did by using the χ2 and unpaired Student t tests. A multiple logistic regression model was used to evaluate independent predictors of visualization of renal medullary hyperattenuation.

Results

Renal medullary hyperattenuation was seen less commonly in patients who received furosemide (27 of 111, 24%) than in those who did not receive furosemide prior to imaging (79 of 178, 44%, P = .001). Bladder urine attenuation was lower in patients who received furosemide (−0.1 HU) compared with those who did not (6.4 HU, P < .001). A multiple logistic regression model revealed independent associations between the visualization of renal medullary hyperattenuation and the absence of furosemide administration (P = .002), younger age (P < .001), and presence of renal stones (P = .047).

Conclusion

Furosemide administration prior to unenhanced CT is associated with decreased visualization of renal medullary hyperattenuation.

© RSNA, 2010

Introduction

A common imaging finding seen at unenhanced computed tomography (CT) is the visible hyperattenuation of the renal medulla when compared with the cortex. This finding has previously been described as a variant of normal kidney appearance (1) and may be related to hydration status (2), owing to the precipitation of drugs in the collecting tubules (3), and a possible indicator of nephrocalcinosis (4). A more recent study demonstrated a correlation between the presence of renal medullary hyperattenuation at unenhanced CT and the urine specific gravity, and suggested that the renal medullary hyperattenuation can be owing to high medullary sodium chloride (NaCl) concentration (5). It is known that loop diuretics such as furosemide are potent inhibitors of the Na-K-2Cl transporter in the renal medulla and dissipate the renal medullary NaCl concentration gradient (6). The administration of furosemide prior to unenhanced CT scans may therefore have an effect on the visualization of renal medullary hyperattenuation.

Patients with complaints of hematuria are frequently evaluated by using CT urography, which includes unenhanced, nephrographic, and excretory phases. Intravenous furosemide is increasingly used prior to CT urography as a strategy to improve ureteral distension and opacification (7–9). Thus, we undertook this study to retrospectively evaluate whether visualization of renal medullary hyperattenuation at the unenhanced phase of CT urography is affected by the administration of furosemide.

Materials and Methods

Patients

This retrospective single-institution study was approved by our institutional review board and compliant with the Health Insurance Portability and Accountability Act; the requirement for written informed consent was waived. An electronic patient information database search was performed to identify all patients over the age of 18 years who underwent CT urography at our institution for evaluation of hematuria from September 2006 to May 2008. Prior to July 2007, CT urography was performed without furosemide. Starting July 2007, all CT urography was performed 20 minutes following the intravenous administration of 10 mg of furosemide (Lasix; Hospira, Lake Forest, Ill). The onset of the diuretic effect of furosemide is 5 minutes, with maximal diuretic effect occurring at 30 minutes after intravenous injection (10).

A total of 305 patients were identified, 16 of which were excluded from analysis owing to the presence of ureteral obstruction (n = 9), extensive CT imaging artifact (n = 5), polycystic kidneys (n = 1), and severely atrophic kidneys (n = 1). Patients with ureteral obstruction were excluded because obstruction is thought to reduce the renal medulla NaCl concentration (5). The remaining 289 patients (152 men, 137 women; mean age, 59 years ± 17 [standard deviation]) were divided in two groups on the basis of whether they did not (n = 178) or did (n = 111) receive furosemide prior to CT. The group of 178 patients who did not receive furosemide comprised 82 women (mean age, 54 years ± 17; range, 18–85 years) and 96 men (mean age, 60 years ± 16; range, 23–87 years). The group of 111 patients who received furosemide comprised 55 women (mean age, 61 years ± 18; range, 29–98 years) and 56 men (mean age, 62 years ± 16; range, 25–89 years). Of all 289 patients, four who did not receive furosemide and one who did had only one kidney. The indication for imaging for all patients was evaluation of gross or microscopic hematuria. Findings seen on CT urograms included renal cell carcinoma (n = 3), bladder transitional cell carcinoma (n = 7), renal cysts (n = 27), nonobstructing renal stones (n = 42), and nonobstructing bladder stones (n = 1). Normal CT scans were obtained from 215 patients. Preimaging serum creatinine was available in 130 (73%) of 178 patients who did not receive furosemide and 80 (72%) of 111 patients who did. There was no significant difference in mean serum creatinine level between the patients who did not receive furosemide and those who did (0.90 vs 0.92 mg/dL [79.56 vs 81.328 μmol/L], P = .44).

CT Technique

All patients were scanned with a 64- (n = 163), 16- (n = 69) or four- (n = 57) section multidetector–row CT scanner (Lightspeed; GE Healthcare, Milwaukee, Wis). As part of the CT urogram protocol, initial unenhanced images were obtained from the diaphragm to the symphysis pubis with a section thickness of 5 mm. A 150-mL bolus of iohexol (Omnipaque 350; Nycomed-Amersham, Princeton, NJ) was then administered at a rate of 2 mL/sec. After a 90-second delay, contiguous 2.5-mm images were acquired from the diaphragm to the symphysis pubis in the nephrographic phase. This was followed by 2.5-mm images from the diaphragm to the symphysis pubis at a 10-minute delay. Images were acquired during a single breath hold for each phase; with a gantry rotation time of 0.8 second; a tube voltage of 120 kVp; and an automated modulation of tube current with image noise set to 12 HU, which was the manufacturer preset for routine abdominal scans. For unenhanced CT, the mean CT dose index was 8.6 mGy and the mean dose-length product was 382.7 mGy·cm. No oral contrast material was administered for these scans.

Image Interpretation

Two radiologists (Z.J.W. and B.M.Y., with 10 and 6 years in abdominal/pelvic CT, respectively) reviewed all unenhanced CT images by consensus on a picture archiving and communication system workstation (IMPAX, version 4.5; Agfa, Mortsel, Belgium) with a level of 40 HU and a window of 80 HU. The two readers were blinded to the date of the examination and whether the patient received furosemide prior to CT. For all 289 examinations, the two readers recorded the presence, distribution, and degree of renal medullary hyperattenuation for each kidney. Medullary hyperattenuation was considered to be present when at least 25% of the medullary pyramids in each kidney were visually of higher attenuation than that of the surrounding renal parenchyma. The distribution of medullary hyperattenuation for each kidney was categorized qualitatively as segmental if less than one-half (<50%) of the medullary pyramids were visually of higher attenuation than the surrounding renal parenchyma, or diffuse if one-half or more (≥ 50%) of the medullary pyramids were visually of higher attenuation than the surrounding renal parenchyma. The degree of medullary hyperattenuation seen in each kidney was qualitatively assessed as faint or obvious. Bladder urine attenuation was recorded for patients by drawing the largest possible elliptical region of interest on the center of the bladder, with care to avoid the bladder wall or areas with obvious artifact. Seven of 289 patients had streak artifact across the bladder from orthopedic hardware in the hip; three received furosemide prior to imaging and four did not. These patients were excluded from the comparison of mean bladder urine attenuation. The size of the regions of interest in the bladder ranged from 8 to 20 cm2. For all patients, the presence of renal and/or ureteral stones was recorded for each kidney by both readers by consensus. Punctate foci of medullary hyperattenuation were considered to be a result of stones and were not considered as renal medullary hyperattenuation.

Statistical Analysis

Statistical analysis was performed by using software (Stata, version 8.0; Stata, College Station, Tex). For the purpose of statistical analysis, renal medullary hyperattenuation was considered present for a particular patient if either kidney demonstrated medullary hyperattenuation; the distribution of medullary hyperattenuation was considered diffuse if either kidney demonstrated diffuse pattern of hyperattenuation; and the degree of medullary hyperattenuation was considered obvious if either kidney demonstrated obvious medullary hyperattenuation.

The comparison of patient demographics and findings between patients who did not receive furosemide prior to CT imaging and those who did was performed by using the following tests: χ2 tests were used to compare the visualization of renal medullary hyperattenuation, frequency of nonobstructing renal stones, and patient sex between the furosemide groups; unpaired Student t tests were used to compare the bladder urine attenuation and mean patient age between the two groups.

We also performed another analysis of all 289 patients comparing the patient demographics and CT findings between patients in whom renal medullary hyperattenuation was visualized at unenhanced CT and those in whom it was not. The χ2 test was used to compare patient sex, the presence of furosemide administration, and the presence of nonobstructing stones between the two groups. An unpaired t test was used to compare the mean patient age and bladder urine attenuation between the two groups.

To assess factors that are independently associated with the visualization of renal medullary hyperattenuation, a multiple logistic regression model was performed by using patient sex, age, the administration of furosemide, and the presence of renal stones as independent variables. For all tests, a P value of less than .05 was considered to indicate a significant difference.

Results

Table 1 shows the patient demographics and findings on CT scans in patients who did not receive intravenous furosemide prior to unenhanced CT and in those who did. Renal medullary hyperattenuation (Figs 1–3) was observed less commonly in scans in which furosemide was administered prior to scanning (27 of 111, 24%) than when it was not (79 of 178, 44%; P = .001) (Fig 4). The distribution of renal medullary hyperattenuation was less diffuse when furosemide was administered (five of 111, 5%) than when it was not (38 of 178, 21%; P < .001) (Fig 4). The degree of renal medullary hyperattenuation was less obvious when furosemide was administered (five of 111, 5%) than when it was not (24 of 178, 13%; P = .013). The mean bladder urine attenuation was lower in scans performed after furosemide administration (−0.1 HU) compared with those without (6.4 HU; P < .001). The mean age, sex, and frequency of finding nonobstructing renal stones were not significantly different between patients who received furosemide prior to CT imaging and those who did not.

Table 1.

Comparison of Patients Who Did Received Intravenous Furosemide prior to Unenhanced CT

graphic file with name 10091769t01.jpg

Open in a new tab

Note.—Unless otherwise noted, data are raw numbers, percentages are in parentheses.

*

Data are the mean ± standard deviation.

†

Indicates significance.

Figure 1a:

Figure 1a:

Open in a new tab

Unenhanced axial CT scans of 20-year-old woman with hematuria. More than one-half of renal medullary pyramids (arrows) in both kidneys (a, upper poles; b, lower poles) have visibly higher attenuation than surrounding renal parenchyma. Renal medullary hyperattenuation was qualitatively described as diffuse in distribution and obvious in degree for both kidneys.

Figure 3:

Figure 3:

Open in a new tab

Unenhanced axial CT scan in 41-year-old woman with hematuria. Renal medullary pyramids are of similar attenuation compared with surrounding renal parenchyma. There was no visualized renal medullary hyperattenuation in either kidney.

Figure 4:

Figure 4:

Open in a new tab

Graph of renal medullary hyperattenuation in patients who did and did not receive furosemide prior to CT examination. Renal medullary hyperattenuation was seen less frequently in patients who received furosemide compared with those who did not. Also, renal medullary hyperattenuation was less diffuse in patients who received furosemide.

Figure 1b:

Figure 1b:

Open in a new tab

Unenhanced axial CT scans of 20-year-old woman with hematuria. More than one-half of renal medullary pyramids (arrows) in both kidneys (a, upper poles; b, lower poles) have visibly higher attenuation than surrounding renal parenchyma. Renal medullary hyperattenuation was qualitatively described as diffuse in distribution and obvious in degree for both kidneys.

Figure 2:

Figure 2:

Open in a new tab

Unenhanced axial CT scan in 49-year-old man with hematuria. Several renal medullary pyramids (arrows) in both kidneys have visibly higher attenuation than surrounding renal parenchyma. Renal medullary hyperattenuation was qualitatively described as segmental in distribution and obvious in degree for both kidneys.

Table 2 shows the patient demographics and CT findings between patients in whom renal medullary hyperattenuation was visualized at unenhanced CT and those in whom it was not. Renal medullary hyperattenuation was seen in 106 (37%) of 289 patients. The mean age of patients with renal medullary hyperattenuation was lower than of those without (53 vs 62 years; P < .001). The mean bladder urine attenuation was higher in patients with renal medullary hyperattenuation than without (5.8 vs 2.7 HU; P = .01). Patient sex and frequency of finding nonobstructing renal stones were not significantly different between the two groups.

Table 2.

Comparison of Patients with Renal Medullary Hyperattenuation Seen at Unenhanced CT

graphic file with name 10091769t02.jpg

Open in a new tab

Note.—Unless otherwise noted, data are raw numbers, percentages are in parentheses.

*

Data are the mean ± standard deviation.

†

Indicates significance.

A multivariate logistic regression model revealed significant independent associations between the presence of renal medullary hyperattenuation at unenhanced CT and the absence of furosemide administration (odds ratio, 2.36; P = .002), younger age (odds ratio, 1.38 [per decade younger age]; P < .001), and presence of renal stones (odds ratio, 2.13; P = .047). Patient sex was not associated with the presence of medullary hyperattenuation (P = .65).

Discussion

We found that the administration of furosemide 20 minutes prior to an unenhanced CT scan decreased the frequency of renal medullary hyperattenuation at unenhanced CT. Furthermore, the medullary hyperattenuation was less diffuse in distribution and less obvious in degree when furosemide was used than when it was not. The findings are likely related to a reduction in renal medullary NaCl concentration as a result of furosemide administration.

The high renal medullary NaCl concentration, which enables the formation of concentrated urine, is a product of the countercurrent exchange mechanism in the nephron. In this countercurrent exchange, the Na-K-2Cl transporters in the ascending limb of the loop of Henle in the medulla actively reabsorb NaCl from the plasma filtrate to the interstitium to generate a high medullary NaCl concentration allowing water reabsorption as a means to maintain fluid homeostasis (11). Pharmacologic reduction of the renal medullary NaCl concentration can be achieved by using furosemide, which inhibits the renal medullary Na-K-2Cl transporter, thereby promoting diuresis (6).

Previous phantom experiments have shown that NaCl solutions at concentrations similar to those in the loop of Henle were radiopaque at unenhanced CT by using commonly used scanning parameters (5). Our findings support the prior observation that high physiologic NaCl concentration in the renal medulla can be visualized at unenhanced CT (5).

Results from a previous study have also shown that bladder urine attenuation correlates with urinary specific gravity, which is a measure of urine concentration (5). Our finding of lower bladder urine attenuation in patients who received furosemide than in those who did not is consistent with the expectation that patients undergoing pharmacologic diuresis will produce more dilute urine.

Our analysis comparing patients with and without renal medullary hyperattenuation at CT, irrespective of furosemide administration, revealed that renal medullary hyperattenuation was independently associated with younger patient age. This finding is consistent with prior work that showed reduced activity of Na-K-2Cl transporters in kidneys of older patients compared with those of younger patients (12). It would follow that with advancing age, the less-active transporters are less likely to generate high medullary NaCl concentration, leading to decreased visualization of renal medullary hyperattenuation at CT in older patients.

While our univariate analysis did not show a relationship between the presence of nonobstructing renal stones and a higher frequency of renal medullary hyperattenuation at CT, our multivariate model did suggest such a relationship. This finding is not surprising because poor hydration is considered an important risk factor for renal stone formation (13,14). In the setting of poor hydration, the kidneys play a crucial role in fluid conservation by maintaining high medullary NaCl concentration to generate highly concentrated urine, which ultimately may lead to higher rates of renal stone formation. Therefore, the multivariate analysis finding of an association between renal medullary hyperattenuation and nonobstructed stones can be explained by the likely higher medullary NaCl concentration in patients with nonobstructed stones.

It is possible that in some patients, renal medullary hyperattenuation may be, in part, a result of microcalcifications, which are radiodense and which have been previously reported in patients at microradiography (15) and histologic examination (16). The microcalcifications have been suggested as precursors to renal stones (15). Two recent studies (17,18) have reported that the attenuation of the renal papillae seen at CT was increased in renal stone formers compared with matched controls. However, these studies did not determine whether the increased attenuation of the renal papillae was a result of microcalcifications, a high NaCl concentration, or other cause. While more work is clearly needed to address this issue, we believe our finding of renal medullary hyperattenuation can largely be attributed to high NaCl concentration because pharmacologic reduction of the NaCl concentration by using furosemide significantly reduced the frequency and prominence of renal medullary hyperattenuation at CT in our study. Also, it is well known that renal stones are associated with older patient age (19). If microcalcifications were the major source for renal medullary hyperattenuation, one would expect it to be associated with older patient age, but our results showed the opposite.

Our study had several limitations. First, all patients in our study were evaluated for hematuria. Future study will be needed with a broader range of patients to determine the relevance of finding renal medullary hyperattenuation in patients with specific renal diseases. Second, because our patients had hematuria, it is possible that hemorrhage in the renal parenchyma or urine may have mimicked or caused renal medullary hyperattenuation or increased bladder attenuation at unenhanced CT in a small percentage of the patients evaluated in our study. Third, while we attempted to control for the most significant possible confounders in our analysis, it is possible that other patient factors, such as hydration status or chronic medication use, may have been different between the groups. Also, because we did not record the exact time of furosemide administration before the unenhanced CT scan, it is possible that the duration between the two may differ slightly between each patient. Future, more rigorous prospective studies would be needed to confirm our findings and to determine the value of unenhanced CT assessments of renal medullary hyperattenuation in specific clinical settings.

Notwithstanding these limitations, we have shown that furosemide administration prior to unenhanced CT is associated with decreased visualization of renal medullary hyperattenuation, which is likely related to the pharmacologic reduction of medullary NaCl concentration.

Advances in Knowledge.

  • •. 

    Renal medullary hyperattenuation at unenhanced CT was seen less commonly in patients who received furosemide (27 of 111, 24%) than in those who did not (79 of 178, 44%) prior to imaging.

  • •. 

    Visualization of the renal medullary hyperattenuation is independently associated with the absence of furosemide administration (P = .002), younger age (P < .001), and presence of renal stones (P = .047).

Implication for Patient Care.

  • •. 

    CT may potentially provide a readily available noninvasive means to monitor changes in renal medullary sodium chloride concentration.

Received September 22, 2009; revision requested October 20; revision received October 31; accepted November 6; final version accepted Novmeber 19.

Authors stated no financial relationship to disclose.

References

  • 1.Lautin EM, Scheiner JE, Rozenblit A, Kaplan D, Frankel-Tiger R, Friedman AC. Differential density of normal renal parenchyma on nonenhanced CT: a new observation. J Comput Assist Tomogr 1996;20(4):616–619. [DOI] [PubMed] [Google Scholar]
  • 2.Tublin ME, Tessler FN, McCauley TR, Kesack CD. Effect of hydration status on renal medulla attenuation on unenhanced CT scans. AJR Am J Roentgenol 1997;168(1):257–259. [DOI] [PubMed] [Google Scholar]
  • 3.Starinsky R, Barr J, Lushkov G, Segal M, Manor A, Golik A. CT of renal densities caused by intravenous infusion of antibiotics. J Comput Assist Tomogr 1995;19(2):228–231. [DOI] [PubMed] [Google Scholar]
  • 4.Curry NS, Gordon L, Gobien RP, Lott M. Renal medullary “rings”: possible CT manifestation of hypercalcemia. Urol Radiol 1984;6(1):48–50. [DOI] [PubMed] [Google Scholar]
  • 5.Hsu CT, Wang ZJ, Yu AS, et al. Physiology of renal medullary tip hyperattenuation at unenhanced CT: urinary specific gravity and the NaCl concentration gradient. Radiology 2008;247(1):147–153. [DOI] [PubMed] [Google Scholar]
  • 6.Cannon PJ, Kilcoyne MM. Ethacrynic acid and furosemide: renal pharmacology and clinical use. Prog Cardiovasc Dis 1969;12(1):99–118. [DOI] [PubMed] [Google Scholar]
  • 7.Roy C, Jeantroux J, Irani FG, Sauer B, Lang H, Saussine C. Accuracy of intermediate dose of furosemide injection to improve multidetector row CT urography. Eur J Radiol 2008;66(2):253–261. [DOI] [PubMed] [Google Scholar]
  • 8.Sanyal R, Deshmukh A, Singh Sheorain V, Taori K. CT urography: a comparison of strategies for upper urinary tract opacification. Eur Radiol 2007;17(5):1262–1266. [DOI] [PubMed] [Google Scholar]
  • 9.Silverman SG, Akbar SA, Mortele KJ, Tuncali K, Bhagwat JG, Seifter JL. Multi-detector row CT urography of normal urinary collecting system: furosemide versus saline as adjunct to contrast medium. Radiology 2006;240(3):749–755. [DOI] [PubMed] [Google Scholar]
  • 10.Furosemide Official FDA information, side effects, and uses. Packet Insert. Revised September 2006 http://www.drugs.com/pro/furosemide-injection.html Accessed September 8, 2009.
  • 11.Rose B, Post T. Clinical physiology of acid-base and electrolyte disorders. 5th ed New York, NY: McGraw-Hill, 2001. [Google Scholar]
  • 12.Lindeman RD, Goldman R. Anatomic and physiologic age changes in the kidney. Exp Gerontol 1986;21(4-5):379–406. [DOI] [PubMed] [Google Scholar]
  • 13.Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A. Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. J Urol 1996;155(3):839–843. [PubMed] [Google Scholar]
  • 14.Pak CY, Sakhaee K, Crowther C, Brinkley L. Evidence justifying a high fluid intake in treatment of nephrolithiasis. Ann Intern Med 1980;93(1):36–39. [DOI] [PubMed] [Google Scholar]
  • 15.Bruwer A. Primary renal calculi: Anderson-Carr-Randall progression?. AJR Am J Roentgenol 1979;132(5):751–758. [DOI] [PubMed] [Google Scholar]
  • 16.Randall A. Papillary pathology as a precursor of primary renal calculus. J Urol 1940;44:580–589. [Google Scholar]
  • 17.Bhuskute NM, Yap WW, Wah TM. A retrospective evaluation of Randall’s plaque theory of nephrolithiasis with CT attenuation values. Eur J Radiol 2009;72(3):470–472. [DOI] [PubMed] [Google Scholar]
  • 18.Eisner BH, Iqbal A, Namasivayam S, et al. Differences in computed tomography density of the renal papillae of stone formers and non-stone-formers: a pilot study. J Endourol 2008;22(10):2207–2210. [DOI] [PubMed] [Google Scholar]
  • 19.Curhan GC. Epidemiology of stone disease. Urol Clin North Am 2007;34(3):287–293. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Radiology are provided here courtesy of Radiological Society of North America

RESOURCES