Patient safety is not compromised by excluding microscopic examination of negative urine dipstick

Abstract

Background

Microscopic examination of samples with negative dipstick results is not necessary. The aims of this study were to: (i) assess the risk of excluding urine sediment examination with negative dipstick results and (ii) calculate time savings by introducing this process.

Methods

The risk analysis was done for samples with negative urine dipstick and positive sediment findings. Possible missed elements in sediment were defined as 21 errors. Time saving was calculated as average time for preparation and examination sediments. Data were presented as counts and percentages.

Results

Out of 2997 samples, negative dipstick results were reported for 926 (30.6%) samples, out of which, microscopic examinations were positive for 527 (17.6%) samples. 18/21 errors were detected, with missing <5 squamous epithelial cells (SQEC) and bacteria 1+ as the most frequent ones (22.7% and 22.4%, respectively). Errors with the intermediate risk for patients were missing to report: ≥5 SQEC, ≥5 transitional epithelial cells, ≥10 hyaline casts (11.9%, 0.21%, 0.32%, respectively). Errors associated with high risk were not detected. Estimated total time saving is more than 25 h/month.

Conclusions

Microscopic examination of urine samples with negative dipstick results can be excluded without risk for patients and can result with considerable time savings.

Keywords

Laboratory optimization patient safety risk analysis urinalysis

Introduction

Urinalysis is composed of physical, chemical and microscopic examination of urine. Urinalysis is very common in diagnosis of renal diseases and urinary tract infections (UTIs) as a screening test, but it is not recommended in asymptomatic patients. Microscopic examination of urine sediment should be performed only if results for physical and chemical examination are abnormal, if microscopy is requested by the physician or in special occasions defined by each laboratory.¹ According to the European urinalysis guidelines, it is not necessary to examine urine samples microscopically if visual examination has normal results (bright yellow and clear urine) and chemical examination is negative on urine dipstick test for proteins, leukocyte esterase, haemoglobin and nitrite.²

Even though, according to guidelines, microscopic examination is not necessary for normal and negative chemical test results, several factors can cause false-negative results and consequently some potentially harmful state can be missed. False-negative urine dipstick results could be caused by the vitamin C intake or dilution of urine samples.³ In spite of that, negative chemical test results have very high negative predictive values for leukocyte esterase and/or nitrite, glucose and haemoglobin (from 70% to even 100%), which is an evidence that it is not necessary to examine sediment microscopically.^4,5 The most of the previously published papers deal with false-positive results. The meta-analysis of Devillé et al. has shown that urine chemical test could be used for exclusion of infection in patients with negative nitrite and leukocyte esterase test. On the other hand, positive results of one or both above mentioned test have to be confirmed with gold standard microbiological urine culture, for diagnosis of UTI.⁶

In the University Department of Chemistry of Medical School University Hospital Sestre Milosrdnice, it is still practice to microscopically examine all urine sediments regardless of results of physical and chemical examination, in order not to miss any potentially pathological element in the sediment. This practice is time consuming for laboratory personnel and can prolong turn-around time. We hypothesize that exclusion of microscopic examination of negative dipstick urine samples would not harm patients or change their management, diagnosis or treatment.

Therefore, the objectives of this study were to: (i) assess the risk for the patients and (ii) calculate the laboratory personnel time savings associated with exclusion of microscopic examination for samples with negative urine dipstick results.

Materials and methods

Study design

All urine samples with requested urinalysis in September 2015 were included in this retrospective study. Data on urinalysis results and requesting departments were collected from the laboratory information system (LIS) of University Department of Clinical Chemistry in Medical School University Hospital Sestre Milosrdnice, Zagreb, Croatia.

Methods

Physical, chemical and microscopic examination of urine samples were performed by experienced laboratory personnel. Urine samples were delivered in the laboratory in urine cups or tubes (Greiner Bio-One GmbH, Kremsmünster, Austria). Combur-10 Test® strips (Roche Diagnostics, GmbH, Mannheim, Germany) were used for chemical examination of urine. Urine analyzers Cobas u411 and Miditron Junior II (both Roche Diagnostics, GmbH, Mannheim, Germany) were used as semi-automated urinalysis system for test strips scanning. Tests were performed as described in manufacturer’s instruction.

For sediment preparation, 10 mL of urine was centrifuged at 500 × g for 10 min on Hettich Rotofix 32 A benchtop centrifuge (Hettich Lab Technology, Beverly, United States). After discarding supernatant, the sediment was resuspended in 500 µL of urine, dropped on glass slide and covered by coverslip. All microscopic examinations were performed by the bright-field microscopy on native urine sediment. Firstly, examination was done under lower magnification of 100 × and fields were scanned for casts. Afterwards, 400 × magnification was used for examination of other elements in urine on 10 random fields.

Risk analysis

All urinalysis results were collected from LIS. Samples with positive dipstick results for proteins, haemoglobin, leukocyte esterase or nitrite were excluded from further analysis, since sediment examination has to be done for these samples. Samples with both negative dipstick and sediment results were also excluded. Study included only samples with negative physical and dipstick results, but with urine sediment results above predefined cut-off values (Table 1). In order not to miss any element in urine sediment, the cut-off values for sample inclusion were set at minimum number of possible found elements in sediment. Amorphous crystals and mucus found in sediment were considered as negative results.

Table 1.

Cut-off values for inclusion of samples in risk analysis.

Elements in urinary sediment/field	Cut-off values for inclusion in the risk analysis
Leukocytes	2
Leukocytes in clumps	1
Erythrocytes	5
Squamous epithelial cells	5
Transitional epithelial cells	1
Hyaline casts	1
Other casts	1
Bacteria	Negative
Crystals (calcium oxalate, triple phosphate, urate, cystine, tyrosine, leucin)	Any amorphous crystals
Yeasts	Negative
Mucus	Any

Based on our extensive literature research, there was no universally defined code and error system for classification of microscopy examination errors. Therefore, in order to be able to quantify our results, we have developed our own scoring system according to the clinical relevance of the specific sediment parameter.

Initially, each author provided their own list of errors. The results were then compared and where differences were observed, reevaluation was done. At the end, the list of errors was approved and agreed on by both authors.

For the risk analysis, 21 potential errors have been identified. Each error was defined as possible missed elements in sediment of negative dipstick urine results on a semi-quantitative scale. Errors from E01 to E03 refer to the missed leukocytes, E04–E06 to the number of erythrocytes, E07–E09 to squamous epithelial cells, E10–E11 to transitional epithelial cells, E12–E15 to casts, E16–E17 to bacteria, E18–E19 to crystals and E20–E21 to the possible missed yeasts. Based on the type and number of missed elements, each of these errors corresponded to certain level of potential harm for the patients. Four different categories of severity for the patients were assessed by the consensus opinion of both authors regarding the clinical significance of elements in urine sediment. The lowest severity level (S1) was defined as minimal harm for patients caused probably by contamination of sample; S2 was defined as necessity for repeated sampling without additional harm for the patient; S3 level as delayed therapy due to missed elements in the urine sediment. Level S4 was defined as the most severe error which corresponds to wrong diagnosis and possible life-threatening state.

Occurrence (O) was classified into five categories according to the observed frequency of each error. O1 category corresponds to the lowest frequency of missed elements (<3%), O2 to the frequency of 3–10%, O3 to >10–25%, O4 to >25–50% and the category of the highest frequency (>50%) of errors was defined as O5.

Risk analysis was done on the basis of ISO 14917 Medical Devices – Application of Risk Management to Medical Devices,⁷ by combining the severity (harm) for the patients (S) and probability of errors occurrence (O) in 4 × 5 matrix. Matrix was divided into three types of coloured fields: green, yellow and red. Errors in green area were considered as low risk and in yellow area as intermediate risk for patients. Red area in matrix was considered to be the most hazardous for patients. In our study, errors in red area with the highest risk were considered unacceptable.

Time saving

Total average time for urine sediment examination was calculated according to the following formula

T_{average} = ((N_{pathological} \times T_{pathological}) + (N_{normal} \times T_{normal})) / N_{total}

where N_pathological was number of samples with pathological sediment results, N_normal number of samples with normal sediment results, N_total total number of urine samples included in the study and T_pathological was average time for examination of 10 pathological urine sediments, T_normal average time for examination of 10 normal urine sediments and T_average total average time for urine sediment examination.

Time saving was calculated as a sum of total average time for urine sediment examination and average time for sediment preparation.

Urine examinations for time saving calculation were done by one experienced laboratory technician. Measurements were done using the stopwatch.

Statistical analysis

Data were presented as count and percentages. Analysis was done in Microsoft Excel (version 2010, Microsoft, USA).

Results

In the studied period (September 2015), 2997 samples with urinalysis requests were admitted to the laboratory. Negative physical and chemical examination results were reported for 926 (30.6%) samples. For 199 (6.6%) Emergency Medicine Department samples with negative dipstick results, urine sediment was not done due to the previous agreement with the Department. Microscopic examination of urine sediment was below predefined cut-off values for 200 (6.7%) samples and those samples were not included in the analysis. For 527 (17.6%) samples, the results of microscopic examination were above cut-off values and risk for patients was assessed for those samples (Figure 1).

Figure 1.

Flowchart of urine samples included into risk analysis. Samples included in the risk analysis are marked in red box, samples excluded from the analysis in grey boxes.

Out of 21 possible errors, 18 errors were found. Corresponding frequencies and assessed severities are presented in Table 2.

Table 2.

Error codes and explanations for possible missed sediment elements in unexamined negative urine with corresponding categories for occurrence according to the assessed severity.

Severity	Error code	Explanation of error, missed elements (N)	Frequency, %	Occurrence category
S1	E01	Leukocytes <5	10.62	O3
	E04	Erythrocytes <5	8.06	O2
	E07	Squamous epithelial cells <5	22.71	O3
	E12	Hyaline casts 2	0.42	O1
	E16	Bacteria 1+	22.40	O3
S2	E02	Leukocytes 5–49	7.99	O2
	E08	Squamous epithelial cells 5–19	11.94	O3
	E17	Bacteria ≥2+	6.10	O2
	E18	Crystals (calcium oxalate, triple phosphate, urate)	3.05	O2
	E20	Yeast 1+	0.53	O1
S3	E03	Leukocytes ≥50	0.21	O1
	E05	Erythrocytes 5-19	2.90	O1
	E10	Transitional epithelial cells < 5	2.10	O1
	E13	Hyaline casts 3-9	0.42	O1
	E21	Yeast ≥2+	0.21	O1
S4	E06^a	Erythrocytes ≥20	0.00	O1
	E09	Squamous epithelial cells ≥20	0.32	O1
	E11	Transitional epithelial cells ≥5	0.21	O1
	E14	Hyaline casts ≥10	0.32	O1
	E15^a	Other casts (urate, granular, renal tubular epithelial, waxy, fatty, red blood cell casts, white blood cell casts)	0.00	O1
	E19^a	Other crystals (cystine, tyrosine, leucin)	0.00	O1

Errors not found and not included in the risk-occurrence table.

Matrix with combination of severity and occurrence categories is shown in Table 3.

Table 3.

Risk-occurrence table with combination of severity and occurrence of error (N = 18).

E: error; O: occurrence; S: severity.

The most frequent errors found in urine sediment were missing <5 squamous epithelial cells (O3 = 22.71%) and bacteria 1+ (O3 = 22.4%). Only four errors with the intermediate risk (yellow field of the matrix) were recorded. Out of errors with intermediate risk, three were due to severity level S4 (missing to report ≥20 squamous epithelial cells, ≥5 transitional epithelial cells and ≥10 hyaline casts with frequencies less than 1% (0.32, 0.21% and 0.32%, respectively) and one error (missing to report 5–19 squamous cells) was due to highest combination of occurrence (O3) and severity (S2). Risk analysis did not identify errors in the red field of 4 × 5 matrix.

The average time for microscopic examination of pathological urine sediment was around 3 min and around 30 s for normal urine samples (Table 4). In the studied period, approximately 20% of samples with negative dipstick results had positive sediment results. When multiplied with proportion of positive and negative sediments, calculated average time for sediment examination was 65 s. Duration of urine sediment preparation was around 1 min per each sample. Average time for sediment preparation and examination was 125 s per sample. For 727 samples with negative dipstick results included in the risk study, 1515 min (around 25 h) or three working days (with 8 h shift) per month could be saved for unnecessary microscopic examination.

Table 4.

Measured time for microscopic examination of 10 pathological and normal urine samples with calculation of total average time for examination and total saving time (in s).

	Time for microscopic examination (s)
Urine number	Pathological urine sediments	Normal urine sediments
1	195	27
2	168	31
3	211	25
4	203	44
5	181	50
6	228	45
7	199	20
8	171	26
9	247	29
10	161	35
Average time per sample (s)	196	33
Proportion of samples^a (N/N_total, %)	146/727 (20)	581/727 (80)
Total average time^b (s)	65
Total saving time^c (s)	125

Distribution of samples with pathological and normal sediment results in the studied period.

Total average time for sediment examination was calculated according to the formula: T_average = ((N_pathological × T_pathological) + (N_normal × T_normal))/N_total.

Time saving was calculated as sum of average time for sediment examination and time for sediment preparation.

Discussion

This study confirms that examination of urine sediment can be excluded if urine has negative dipstick test results for proteins, leukocytes, nitrite and haemoglobin. We have proved that patients are not at higher risk if this procedure is followed. Although guidelines for urinalysis exist for a long time (more than 15 years), this kind of risk analysis for patients with negative dipstick results was never done. Additionally, this approach provides saving time for laboratory personnel without interfering with the quality of the reported results or increasing the harm for patient safety.

Errors with the highest frequency were bacteria in trace (1+) and few (less than 5) epithelial squamous cells in the sediment. Although those errors had the highest frequencies, both were not exceeding 23%. Findings of bacteria and epithelial squamous cells are usually associated with improper urine collection or prolonged storage and transportation to the laboratory on temperature higher than allowable. For examination of bacteria and epithelial cells, urine samples are stable for maximum of 3 h.³ If samples are not transported to the laboratory immediately after collection, commensal bacteria present in urine and urinary tract can grow in samples and false positive results can appear. Moreover, it has been previously published that patients usually do not adhere to instructions for proper urine collection.⁸ which could contribute to the contamination of sample. Patients are usually not informed about all preanalytical requirements (adequate container, delayed delivery to the laboratory and proper genital cleaning for spot urine collection).⁹ Therefore, the most frequently missed elements are probably the result of inadequate preanalytical procedures and are not associated with the UTI. Accordingly, the severity of those errors was defined as the lowest possible (S1), and patient’s health is certainly not at risk if those elements are not reported.

In our study four errors of intermediate risk for patients were found. Out of those four, only one (5–19 epithelial cells) had frequency around 12% and other three (more than 5 transitional epithelial cells, more than 20 squamous epithelial cells and more than 10 hyaline casts) were defined as errors with the highest severity but were found in only eight samples (with frequency of each less than 0.4%). Hyaline casts found in urine sediment can be associated with urine concentration in after-night sample or may be present after strenuous physical activity. Small number of casts in urine usually has no clinical significance. However, finding of hyaline casts may be associated with some renal diseases and special attention has to be dedicated for such results. Cellular casts composed of erythrocytes, leukocytes or epithelial cells could be associated with glomerular or tubular diseases such as glomerulonephritis and pyelonephritis or even with states of higher risk for patients as acute tubular necrosis and even intoxication with nephrotoxic elements.¹⁰ The importance of hyaline and other casts for the diagnosis of renal diseases is the reason why we have chosen such a rigid categorization of estimated risk for patients.

On the other hand, false-positive epithelial cells in urine sediment may be found commonly, especially in women, even without UTI.¹¹ UTI diagnosis may be set after positive nitrite on urine dipstick and/or with confirmation of bacterial culture in urine.¹² We may presume that high number of epithelial cells in our study may be related to poor collection technique without previous cleaning of urogenital area. If symptoms of UTI are present in those patients, physicians would probably ask for repeated urinalysis and patients would be treated appropriately.

Haematuria in the urine could have high clinical significance in diagnosing of primary and secondary glomerulonephritis and also some tubulointerstitial kidney diseases or malignant conditions.⁴ Haematuria can usually be detected with positive dipstick results, and the most important findings are dysmorphic erythrocytes in urine sediment, which imply that glomerular damage is occurring. Although few guidelines exist for managing the patients with microhaematuria, there is no published study with research of negative chemical test results and positive microscopic examination. The most frequent result for erythrocytes found in our study is less than 5 cells per field (E04), which is still within the normal values. Also, no more than 19 missed erythrocytes per examined field under the bright-field microscope were observed in this study. Therefore, we conclude that the risk that severe haematuria is missed, if sediment examination was excluded, is very low.

According to the ISO 15189 standard, all processes in the laboratory should be evaluated for potential risk for the patient safety. If some of laboratory processes are found to be of high risk for patients, they should be changed immediately in order to reduce harm.¹³ The main objective for all laboratory professionals is providing the highest level of patient care, with the lowest error rate and personnel utilization. In order to obtain these goals, laboratory personnel needs to ensure the optimal quality of the preanalytical, analytical and postanalytical phase as well as optimize the entire procedure.^14,15 Errors in process could harm, not only patients, but also laboratory personnel, due to unnecessary waste of time for unproductive processes.¹⁶ Therefore, the analysis of risk for the patients combined with the productivity analysis of the work process is the optimum way of gaining the benefit for the laboratory.

In our laboratory, all urine samples were examined microscopically, except for less than 7% of samples received from Emergency Department. Preparation and examination of urine sediments are very time consuming.¹⁷ The average time spent for microscopic examination of negative dipstick urines was around 2 min, but with large number of samples in daily practice, laboratory personnel can save more than three working days per month. Laboratory personnel included in this part of study has a lot of experience in the microscopic examination of urine and calculated time saving may be even higher if we have included more than one laboratory technician in this estimation. However, time saving with the new procedure is not negligible.

This study has few limitations. Since our results were collected retrospectively, we do not have information about the time of sample collection, storage and transportation prior to urinalysis. Information about the first or later stream of urine was also not available. Furthermore, it is not possible to check if samples of studied patients were contaminated with some other interfering substances such as vitamin C.¹⁸

Conclusions

This risk analysis on exclusion of unnecessary sediment microscopy confirmed that the quality of reported results is not compromised; physicians have access to almost the same clinical information, patients are not at risk of diagnostic errors and productivity within laboratory is increased.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval (including reference number)

Not applicable. Study is retrospective, without info on patient data.

Guarantor

MM.

Contributorship

MM and NN researched literature and conceived the study. MM was involved in data collection. MM and NN have done data analysis. MM wrote the first draft of the manuscript. Both authors reviewed and edited the manuscript and approved the final version of the manuscript.

References

Clinical and Laboratory Standards Institute (CLSI). Urinalysis. Approved Guideline-Third Edition; CLSI Document GP16-A3. Wayne, PA: Clinical and Laboratory Standards Institute, 2009.

European Urinalysis Guidelines; under the auspices of the European Confederation of Laboratory Medicine (ECLM). Scand J Clin Lab Invest 2000; 60: 1–96.

Delanghe

Speeckaert

. Preanalytical requirements of urinalysis. Biochem Med (Zagreb) 2014; 24: 89–104.

Simerville

Maxted

Pahira

. Urinalysis: a comprehensive review. Am Fam Physician 2005; 71: 1153–1162.

Zamanzad

. Accuracy of dipstick urinalysis as a screening method for detection of glucose, protein, nitrites and blood. East Mediterr Health J 2009; 15: 1323–1328.

Devillé

Yzermans

Van Duijn

et al.

The urine dipstick test useful to rule out infections. A meta-analysis of the accuracy. BMC Urology 2004; 4: 4–4.

ISO14971:2012. Medical devices: application of risk management to medical devices. Geneva: ISO, 2012.

Miler

Simundic

. Low level of adherence to instructions for 24-hour urine collection among hospital outpatients. Biochem Med (Zagreb) 2013; 23: 316–320.

Nikolac

Simundic

Kackov

et al.

The quality and scope of information provided by medical laboratories to patients before laboratory testing: survey of the Working Group for Patient Preparation of the Croatian Society of Medical Biochemistry and Laboratory Medicine. Clin Chim Acta 2015; 450: 104–109.

10.

Ringsrud

. Casts in the urine sediment. Lab Med 2001; 32: 153–155.

11.

Frazee

Enriquez

et al.

Abnormal urinalysis results are common, regardless of specimen collection technique, in women without urinary tract infections. J Emerg Med 2015; 48: 706–711.

12.

Najeeb

Munir

Rehman

et al.

Comparison of urine dipstick test with conventional urine culture in diagnosis of urinary tract infection. J Coll Physicians Surg Pak 2015; 25: 108–110.

13.

International Organization for Standardization ISO 15189:2012. Medical laboratories: particular requirements for quality and competence. Geneva: ISO, 2012.

14.

Lippi

Becan-McBride

Behúlová

et al.

Preanalytical quality improvement: in quality we trust. Clin Chem Lab Med 2013; 51: 229–241.

15.

Topic

Nikolac

Panteghini

et al.

How to assess the quality of your analytical method?

Clin Chem Lab Med 2015; 53: 1707–1718.

16.

Njoroge

Nichols

. Risk management in the clinical laboratory. Ann Lab Med 2014; 34: 274–278.

17.

Piccoli

Gai

Mezza

et al.

Biochemical and microscopic urinalysis: time and cost in a nephrology laboratory. J Nephrol 2002; 15: 575–579.

18.

Jeong

Kim

et al.

Influence of vitamin C on urine dipstick test results. Ann Clin Lab Sci 2015; 45: 391–395.