The Accuracy of 2.5 mm Punch Biopsy in theDiagnosis of Skin Lesions

Background: Punch biopsies are gaining widespread attention among medical professionals for their broad use and ease of access. Despite decades-long use, there needs to be more robust statistical evidence regarding their diagnostic accuracy. This study aims to evaluate the accuracy of the 2.5 mm punch biopsy in diagnosing skin lesions by comparing the histopathology of punch biopsies with that of excisional biopsies.

Methods: In this retrospective study, a review of 4,000 charts, including skin lesions seen by a single plastic surgeon from 2016–2023, was conducted to identify patients who underwent a 2.5 mm punch biopsy of a lesion followed by a subsequent excisional biopsy. 206 charts were identified. Concord ance between punch and excisional histopathologies was used to calculate the efficacy of the 2.5 mm punch biopsy as a diagnostic tool.

Results: Of 206, 141 skin lesions were characterized as cancerous by punch biopsy, all confirmed on subsequent excisional biopsy. 51 were deemed benign/precancerous on punch biopsy and confirmed by excision. 12 were identified as cancerous by punch biopsy but later characterized as benign/precancerous by excision. 2 were characterized as precancerous by punch biopsy and cancerous by excision. Analysis revealed that the 2.5 mm punch biopsy had a sensitivity of 98.6% (95% CI: 95.04%–99.83%) and specificity of 80.95% (69%–89.75%) in diagnosing skin lesions, which is statistically significant by several measures.

Conclusion: The 2.5 mm punch biopsy is an accurate tool for diagnosing skin lesions. It can be applied to various anatomical sites and lesion sizes. Its non-suturing requirement may enhance cosmetic outcomes and ease-of-use.

Introduction

A punch biopsy involves the removal of a small cylindrical piece of tissue from the skin or other tissues for microscopic examination and diagnosis [1], [2]. Punch biopsies can be classified as incisional or excisional, depending on the lesion size [3]. Various medical professionals, including plastic surgeons, dermatologists, and family medicine physicians, widely use the punch biopsy [4]. Their widespread popularity is due to versatility, ease, and minimal invasiveness, while maintaining high diagnostic accuracy by providing a small but sufficient amount of tissue for microscopic analysis [5].

The spectrum of skin conditions for which punch biopsies can help diagnose, includes but is not limited to, skin cancers, precancerous skin conditions, as well as benign dermatological skin lesions [6]–[8]. Insight regarding the cause, extent, and severity of a skin condition can all be elucidated by a punch biopsy [6].

Punch biopsies have been utilized for decades with minimal scientific statistical evidence of their accuracy. This study aims to evaluate the accuracy of a ‘smaller’ 2.5 punch biopsy by comparing its diagnostic agreement with an excisional biopsy.

In our practice, after several years of using punch biopsies ranging from 2 to 8 mm, we observed that a 2.5-punch biopsy provides enough tissue, does not require suturing, and is quick and easy to perform. With a smaller sample size, the biopsy must be proven accurate. A review of the existing English literature reveals that no previous studies have assessed the accuracy of a 2.5 punch biopsy. Therefore, this study was conducted to scientifically and statistically address this question. This study aims to evaluate the accuracy of a smaller 2.5 mm punch biopsy by comparing its diagnostic agreement with that of an excisional biopsy.

This paper’s findings are designed to contribute to the literature surrounding the accuracy of smaller-sized punch biopsies in diagnosing skin lesions. Pursuing these objectives, this study seeks to give physicians confidence in choosing smaller punch biopsy sizes, thus leading to more accessible procedures and better cosmetic outcomes for patients.

Materials and Methods

Study Design

This study was designed as a retrospective chart review examining the diagnostic accuracy of a 2.5 mm punch biopsy. The study analyzed patient records from a single plastic surgeon’s outpatient clinic, as part of the regional healthcare system and its associated academic institution. Ethics approval for this retrospective study was obtained from the Hamilton Integrated Research Ethics Board (HIREB). For this study, we defined basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and Malignant melanoma (MM) as cancerous lesions. We defined actinic keratosis and cheilitis as precancerous lesions.

A chart review of 4000 cases corresponding to patients with skin lesions seen by a single plastic surgeon between January 2016 and August 2023 was conducted. Inclusion criteria involved all patients who received a 2.5 mm punch biopsy and a subsequent excisional biopsy with recorded dates of both punch biopsy and excisional biopsy. All punch biopsies were performed under local anesthesia in office settings by a single plastic surgeon. A disposable 2.5 mm punch biopsy (Integra Lifesciences Corp. USA) was applied to most anatomical sites, and for different sizes of skin lesions, two or more punch biopsies were obtained from large lesions. No sutures were used regardless of the site of the biopsy and the size of the lesions. Exclusion criteria include patients referred for radiation therapy post-punch biopsy, patients with prior biopsies of the same lesion, or those who had their punch biopsy done elsewhere. A total of 206 patients who met these criteria were identified.

Data extraction included patient demographics, anatomical site and date of biopsy, and relevant histopathological findings of both punch and subsequent excisional biopsies. The extracted patient data was subsequently de-identified, stored securely and only accessed by authorized personnel who were signees to the research ethics board agreements.

The collected data values were used to perform statistical analyses of the diagnostic accuracy of the 2.5 mm punch biopsy by determining the concordance of the histopathological diagnosis of the punch biopsies with the gold-standard excisional biopsy. This was further assessed by calculating the sensitivity and specificity of punch biopsy relative to excisional biopsy.

To assess the statistical significance of these calculations, Chi-square tests were utilized. These statistical analyses were carried out by our biostatistician using the IBM SPSS Statistics Software package.

Results

A total of 206 charts were used to estimate power using related proportions to detect differences between the two tests in 95% of cases. A power of 0.8 was obtained to reflect a difference of 4% or larger between the two tests.

Descriptive Data

Collected data in this study were stratified by age and histopathological diagnosis. The average age of patients was 70 years old, median age was 72 years old, and range was 9–99 years old. The most common age group among our included patients was 80–89 years old. Eight patients were 90–99 years old. The anatomical site was recorded for each biopsy. We calculated diagnostic metrics, such as sensitivity, specificity, and accuracy, of 2.5 mm punch biopsy relative to excisional biopsy for each anatomical site.

Overall Diagnostic Methods

While we categorized skin lesions into five categories–MM, SCC, BCC, precancerous, and benign–to construct a 2 × 2 confusion matrix, we categorized data into two groups: 1) cancer (MM, SCC, or BCC); and 2) precancerous or benign.

Raw n-values for these groups presented within a confusion matrix are shown in (Fig. 1). Within this confusion matrix, 2.5 mm punch biopsy had a sensitivity of 98.6%, reflecting high efficacy in identifying true positive cases for cancer. A specificity of 80.95% indicates that the biopsy is efficacious in identifying true negative cases but with higher rates of inaccurately identifying lesions as cancerous when they are not. Diagnostic performance metrics for 2.5 mm punch biopsy relative to gold-standard excisional biopsy–based on the groups demonstrated in (Fig. 1)–are summarized in (Table I).

Fig. 1. Raw n-values of the histopathological findings of 2.5 mm punch and excisional biopsy presented as a 2 × 2 confusion matrix

Table I. Diagnostic Metrics for 2.5 mm Punch Biopsy Relative to Excisional Biopsy

Metric	Value [CI]
Sensitivity	98.6% [95.04%–99.83%]
Specificity	80.95% [69.09%–89.75%]

Fig. 1. Raw n-values of the histopathological findings of 2.5 mm punch and excisional biopsy presented as a 2 × 2 confusion matrix.

Overall, across 206 patients, there was a 93.20% accuracy of 2.5 mm punch biopsy relative to excisional biopsy within a binary system of determining cancer versus precancerous or benign. Pearson Chi-square between 2.5 mm punch biopsy histopathological results and excisional biopsy results was statistically significant at p = 0.000; Cohen’s Kappa coefficient, after ruling out random error, was ‘almost in perfect agreement’ at 82% and also statistically significant at p = 0.000 (see Table II) [9].

Table II. Statistical Tests on Overall Diagnostic Metrics of a 2.5 mm Punch Biopsy

Chi-square tests for 2.5 mm punch biopsy relative to excisional biopsy
Test		Value		df	Asymptotic significance (2-sided)
Pearson chi-square		256.660		4	0.000
Symmetric measures for 2.5 mm punch biopsy relative to excisional biopsy
Method	Measure	Value	Asymptotic standard errora	Approximate Tb	Approximate significance
Measure of agreement	Cohen’s Kappa	0.822	0.042	14.777	0.000

Diagnostic Metrics by Disease Category

Returning to a binary system, we examined a 2.5 mm punch biopsy’s ability to discriminate diagnoses among the different histopathological diagnoses of cancer. n-values demonstrating the frequency of histopathological diagnosis across 206 lesions biopsied, both punch and excision, are shown in Table III. For each histopathological diagnosis, data relating to the confusion matrices (TP, FP, FN, and TN values) is also included and was used for the calculation of sensitivity, specificity, and accuracy, as shown in Table IV.

Table III. Raw values of each Histopathological Diagnosis and Confusion Matrices Values

		Excision
		BCC (n = 75)	SCC (n = 50)	Melanoma (n = 18)	Benign (n = 47)	Pre-cancerous (n = 16)	FP
Biopsy	BCC (n = 75)	73	1	0	1	0	2
	SCC (n = 59)	2	47	0	5	5	12
	Melanoma (n = 19)	0	0	18	1	0	1
	Benign (n = 38)	0	0	0	38	0	0
	Pre-cancerous (n = 15)	0	2	0	2	11	4
	FN	2	3	0	9	5

Table IV. Diagnostic Metrics for 2.5 mm Punch Biopsy Relative to Excisional Biopsy Categorized by Cancer Type

Metric		Value
	BCC	SCC	Melanoma
Sensitivity	97.33% [90.70%–99.68%]	94.00% [83.45%–98.75%]	100.00% [81.47%–100.00%]
Specificity	98.47% [94.59–99.81%]	92.31% [86.95%–95.65%]	99.47% [97.07%–99.99%]
Accuracy	98.06% [95.10%–99.47%]	92.72% [88.27%–95.87%]	99.51% [97.33%–99.99%]

Discussion

The 2.5 mm punch biopsy strikes a careful balance between collecting enough tissue for histopathological analysis and minimizing patient discomfort. This procedure aims to maximize tissue collection while also ensuring ease of performance and maintaining patient comfort. Additionally, it provides a sufficient tissue sample to achieve statistically significant diagnostic accuracy compared to the gold standard and more invasive excisional biopsy methods.

The 2.5 mm punch biopsy can be considered a reliable tool in determining if a lesion is cancerous or noncancerous. Of 206 lesions biopsied and then excised, 143 were ultimately determined to be cancerous on excisional biopsy, and all but two had previously been identified as cancerous on 2.5 mm punch biopsy. These results provided a sensitivity of 98.6%. The specificity of the 2.5 mm punch biopsy was less impressive than its sensitivity but was still acceptable at 80.95%. However, unnecessary further testing due to false positives may occur as a result of slightly lower specificity. One highlight of our paper was that 2.5 mm punch biopsy appeared to preferentially operate on the side of greater sensitivity than specificity for cancer diagnoses. While over-investigating due to a test having lower specificity has drawbacks, the ramifications of missing a cancer diagnosis due to a test having poor sensitivity can be dire [9]. Additionally, concerns about excessive investigation are alleviated by the fact that a 2.5 mm punch biopsy is associated with lower morbidity. This procedure can serve as a less invasive and quicker alternative to a more extensive excisional biopsy when there is uncertainty regarding the clinical diagnosis. During our chart review we found several cases where lesions that were highly suspicious for melanoma based on clinical examination were immediately excised. In contrast, punch biopsies were only conducted on lesions with a low suspicion for melanoma on clinical examination [10].

The 2.5 mm punch biopsy has shown impressive sensitivity and overall accuracy for diagnosing the various types of skin cancer. The highest overall accuracy was found for melanoma, followed by basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). This study demonstrates that the 2.5 mm punch biopsy is an effective tool for diagnosing skin cancers, particularly melanoma. Notably, no case of melanoma was missed by the punch biopsy. The ability of the 2.5 mm punch biopsy to effectively detect melanoma is crucial, as timely diagnosis is essential for initiating treatment more quickly.

These findings encourage clinicians to choose the smaller 2.5 mm punch confidently.

The common belief that it may miss cancer cells due to inadequate tissue is unfounded. Our research shows that the 2.5 mm biopsy provides sufficient tissue samples, requires less anesthesia, reduces patient discomfort, and offers better cosmetic outcomes compared to larger biopsies.

Conclusion

This study demonstrates that a 2.5 mm punch biopsy is an effective and accurate screening tool with strong sensitivity in identifying cancer. It can be applied to various anatomical sites and lesion sizes. Its non-suturing requirement may enhance cosmetic outcomes and ease-of-use. Statistical analysis revealed that a 2.5 mm punch biopsy yielded 98.6% sensitivity [95% CI of 95.04%–99.83%] and 80.95% specificity [69%–89.75%] in diagnosing skin lesions.

Our study was limited by a relatively small sample size comprising 206 lesions. A significant factor contributing to this limitation was the necessity to review physical copies of certain charts. Furthermore, the unexpected disruption resulting from the COVID-19 pandemic posed challenges for both patients and researchers in accessing health care services. We look forward to future research that will statistically assess the accuracy of different punch biopsy sizes, paving the way for improved diagnostic practices and patient outcomes.