Chapter 17: Surgery Coding I: The Surgical Package, Global Periods, Integumentary, and Musculoskeletal

"The operative note tells you what was done. The guidelines tell you how much of it you already got paid for." — constructed

Overview

Surgical coding is where two things this book has been building finally collide.

The first is the surgical package. A surgical code does not pay for the operation. It pays for a bundle — a period of care around the operation, defined in advance, whose contents you are expected to know without being told. Report something already inside that bundle and the claim is wrong in a way that is invisible on the note and obvious on a remittance.

The second is the operative report. Everything in Part II and most of Part III has been coded from notes that were, by design, orderly. An operative report is not orderly. It is dictated, it is written for surgeons rather than for coders, and the sentence that decides the code is frequently not in the section where you would look for it.

This chapter and the next are the surgery chapters. This one covers the package and global periods — which govern every surgical code in both chapters — and then the two sections that generate the most outpatient surgical volume in the country: integumentary and musculoskeletal.

Three specific things in this chapter are worth the price of the chapter on their own:

• The measurement rule for lesion excision, which almost everyone gets wrong, in a way that costs money on every claim
• The fracture-care decision, which is a genuine fork with real consequences and no default
• The arthroscopy patterns that produced Account 31-2245's \$25,720.80 demand

In this chapter, you will learn to:

• Say what the surgical package includes and what it does not
• Read every global period indicator, and find it in a free file
• Read an operative report in a fixed order
• Measure a lesion excision correctly, and know why the pathology report is the wrong source
• Classify and add repairs
• Tell grafts from flaps, and code debridement by depth
• Make the fracture-care decision
• Report casting and splinting
• Recognize the unbundling pattern before it becomes an audit

17.1 What the surgical package includes, and what it excludes

CPT's surgical package is a set of services always included in a given surgical procedure's payment, in addition to the operation itself.

The specific contents are stated in the CPT Surgery guidelines and you should read them there. In substance, the package covers:

• Evaluation and management services on the date of or immediately before the procedure, after the decision for surgery has been made — including the history, examination, and the process of obtaining consent
• Local anesthesia, digital block, or topical anesthesia
• Immediate postoperative care, including dictating the operative note and talking with the family and other professionals
• Writing orders
• Evaluating the patient in the post-anesthesia recovery area
• Typical postoperative follow-up care

What that phrase actually means for you

The package is why a procedure and an E/M on the same day are a question rather than an assumption. Chapter 14 §14.4 approached this from the modifier side; this is the same rule from the other direction. The evaluation that led to the decision is included. Modifier 25 is asking whether there was work above and beyond it.

The package is also why "typical postoperative follow-up care" is not billable, which is the single most common surprise for practices new to surgical billing: the visit at two weeks, the suture removal, the wound check, and the reassurance are all inside the payment already received.

What is EXCLUDED

Read this list twice. Everything on it is separately reportable, and practices routinely fail to report it — which is Chapter 14's silent underpayment again, in the highest-dollar setting in the book.

The initial consultation or evaluation at which the decision for surgery was made. For a major procedure this is what modifier 57 identifies. It is not in the package. It is a separately payable service and it is written off constantly.

Complications requiring a return to the operating room. Modifier 78 — Chapter 14 §14.9.

Treatment for the underlying condition, or an added course of treatment that is not part of normal recovery.

Unrelated services during the postoperative period — modifier 79 for procedures, modifier 24 for E/M.

Supplies beyond those usually included.

More extensive procedures where a lesser procedure fails and a greater one is required.

⚠️ Where Claims Die

The single largest silent loss in surgical billing is the decision-for-surgery visit.

A surgeon sees a patient in consultation, evaluates them, and decides to operate. That evaluation is not in the package — it is the visit at which the decision was made, and modifier 57 identifies it for a major procedure.

Practices write it off anyway, for a reason that is easy to sympathize with: somebody once got a denial for billing an E/M near a surgery, concluded that E/Ms near surgery are not payable, and made a rule. The rule then outlives everyone who remembers why.

It has no financial signal. Nothing denies, because nothing is submitted. The revenue simply is not there, and no report in the building shows an absence.

Chapter 15 §15.9a said it about G2211 and it is exactly as true here: a code you never report can never be denied.

17.2 Global periods: 000, 010, 090, XXX, YYY, ZZZ

The surgical package has a duration, and the duration is the global period.

Three consequences fall out of this table immediately.

Whether a procedure is "minor" or "major" is not a judgment. It is a lookup. 000 and 010 are minor; 090 is major. This is what decides modifier 25 versus modifier 57 — Chapter 14 §14.4 — and the answer is in a file, not in an opinion about how big the operation felt.

A 090 global includes the day before surgery. Practices forget the preoperative day exists, and it is the source of a specific and confusing denial: an E/M the day before a major surgery, denied, correctly, because it is in the package.

ZZZ is why add-on codes never carry their own global. Chapter 13 §13.7 said add-on codes are always reported with a primary procedure; this is the payment-side reason.

🔢 Code It

Where do you find the global period?

The Medicare Physician Fee Schedule relative value file. It is free, it is published by CMS, and it carries the global period indicator in a column, for every code.

Chapter 14's further reading said four of that chapter's questions were answered by four columns in one free file. This is the first of the four, and it is the one you will use most:

```text LOOK UP THE CODE. READ THE COLUMN.

 20610  →  000   minor, no postoperative days
 11042  →  000
 12002  →  010   minor, 10 postoperative days
 29827  →  090   major
 +29826 →  ZZZ   add-on; lives inside 29827's global

```

A coder who guesses at global periods will be wrong several times a week. A coder who looks them up will not, and the lookup takes eleven seconds.

The global period and the modifiers

Chapter 14 §14.9 built the decision tree. Here is what it is standing on:

• 24 — unrelated E/M during a postoperative period
• 57 — decision for major surgery (090)
• 58 — planned or staged related procedure
• 78 — unplanned return to the operating room
• 79 — unrelated procedure during a postoperative period

None of those modifiers means anything until you know the global period, which is why this section precedes everything else in the chapter.

🗂️ The Encounter — 20610's 000-day global

Account 10-4471's line 2 is 20610-RT, and its global period is 000.

Trace what that means, precisely.

Included, and therefore not separately billable: the physician's evaluation of the knee immediately before the injection after the decision to inject was made; positioning the patient; the local anesthetic — the note documents 1% lidocaine, which is inside the package and is not reported; obtaining consent; performing the injection; the dressing; and post-procedure instructions.

Not included: the significant, separately identifiable E/M — three chronic conditions assessed, prescription drug management, two tests ordered — which is why line 1 exists and carries modifier 25. And the drug itself, methylprednisolone acetate, which is line 3, J1030: a supply, not a service, and outside the package.

And there are no postoperative days at all. A 000-day global has none. If this patient returns in three weeks for a wound check, that visit is not in any package — it is an ordinary E/M, billed normally, with no modifier required.

Which settles two questions students always ask. Why 25 and not 57: 20610 is 000, therefore minor, therefore 25. And why the lidocaine is not on the claim: local anesthesia is in the package, and a practice billing it separately is billing for something it has already been paid for.

17.3 Reading an operative report as a coder

An operative report is not a coding document and it was not written for you. Read it in a fixed order, every time, because the order is what protects you from the report's structure.

   THE ORDER

   1. PREOPERATIVE DIAGNOSIS ....... what they thought
   2. POSTOPERATIVE DIAGNOSIS ...... what they found        ◄ code from THIS
   3. PROCEDURE(S) PERFORMED ....... the surgeon's own list ◄ a claim, not a fact
   4. THE BODY OF THE NOTE ......... what actually happened ◄ the SOURCE OF TRUTH
   5. Findings, specimens, EBL, closure

Three rules follow from that diagram, and they are the whole section.

Code from the postoperative diagnosis, not the preoperative one. The preoperative diagnosis is what was suspected. The postoperative diagnosis is what was found, and it is what the encounter established.

The "Procedure Performed" heading is a summary written by a person, and it is not authoritative. Surgeons write it before dictating, or from a template, or from habit. It is frequently incomplete and occasionally wrong — describing a planned procedure that was modified, omitting something performed, or naming a procedure with a term that does not map cleanly to a code.

The body of the note governs. What was actually incised, excised, measured, repaired, and closed. When the heading and the body disagree, the body wins — and you query rather than choosing.

📋 Read the Chart

Source: operative report What it says:

Procedure performed: "Excision of lesion, left forearm."

Body of note: "…a 1.8 cm pigmented lesion of the left forearm was excised with 0.4 cm margins circumferentially. The resulting defect measured 2.6 cm and was closed in layers, with the deep dermal layer approximated using 4-0 absorbable suture and the skin closed with 5-0 nylon."

What it means: the heading names one procedure. The body contains three separately codeable facts the heading does not mention: the excised diameter including margins (§17.4), a layered closure which is an intermediate repair and separately reportable (§17.5), and the defect size, which is the measurement the repair is coded from.

What to do about it: code from the body. A coder who codes the heading reports one code and misses the repair entirely — and misses the margins, which are what determine the excision code's size tier.

Where it appears: in a very large share of dermatologic and general surgical operative notes. The heading is almost never a lie. It is almost always a summary.

⚖️ Compliance Check

You may not code a procedure that is not documented in the body of the note, and you may not decline to code one that is.

Both halves matter. Coders internalize the first and are inconsistent about the second — a documented procedure that the surgeon did not list is still a documented procedure, and choosing not to report it because the surgeon "didn't ask for it" is a coding decision made on the wrong basis.

The correct move when the heading and the body disagree is a query, not a choice. Chapter 33 §33.5 covers the form. The query is short: "The body of the note documents a layered closure. Should a repair be reported in addition to the excision?" — open, non-leading, and answerable in one word.

17.4 The integumentary system: lesions, excision, and the measurement rule

The integumentary section covers skin, subcutaneous tissue, and accessory structures. It is high-volume, it is heavily audited, and it contains the rule this chapter exists to teach.

Benign versus malignant

Excision codes are divided by whether the lesion was benign or malignant, and by anatomic site, and by size.

The pathology report determines benign versus malignant — which means the code frequently cannot be assigned on the day of surgery. That is normal and correct. A practice that codes excisions before pathology returns is guessing, and it will guess wrong in the expensive direction some of the time.

THE MEASUREMENT RULE

The size is the LESION'S GREATEST DIAMETER PLUS THE NARROWEST MARGINS REQUIRED, measured BEFORE excision.

Not the pathology specimen. Not the defect. Lesion + margins, at the time of excision.

Three things go wrong here, constantly.

The pathology report is the wrong source. Tissue shrinks in formalin. A specimen measured in the laboratory is smaller than the tissue that was excised — sometimes substantially. Coding from the pathology report systematically undersizes every excision, which systematically undercodes and underpays, on every claim, forever, with no signal that anything is wrong.

The margins count, and they are frequently not documented. The rule is lesion diameter plus the narrowest margin, on both sides. A 1.8 cm lesion with 0.4 cm margins is:

   0.4  +  1.8  +  0.4  =  2.6 cm excised diameter
   ───     ───     ───
  margin  lesion  margin

Not 1.8. Not 2.2. 2.6. And that difference frequently crosses a code's size tier.

The defect is not the measurement either. The defect — the hole left behind — is what the repair is measured from (§17.5), not the excision. The two measurements come from the same operation and are not the same number.

🧮 Run the Numbers

One lesion, four measurements, three of them wrong.

A pigmented lesion of the forearm. The surgeon documents a 1.8 cm lesion excised with 0.4 cm margins. The resulting defect measures 2.6 cm. The pathology report describes a specimen measuring 2.3 cm.

Number	What it is	Use it for
1.8 cm	the lesion alone	nothing — it is not the excision size
2.6 cm	lesion + margins	THE EXCISION CODE
2.6 cm	the defect	the repair code, if a repair is reported
2.3 cm	the formalin-shrunken specimen	nothing

The excised diameter and the defect happen to be the same number here, which is common and which is why the two are so easily confused. They are different quantities that happened to agree. Change the closure technique and the defect changes while the excision does not.

And the pathology number is 0.3 cm smaller than reality, which on a busy dermatology practice's claim volume is a permanent, invisible, self-inflicted discount.

🎓 Exam Watch

The measurement rule is on every certification exam, usually with all four numbers present and three of them as distractors.

Say the rule out loud before you look at the answers: lesion diameter plus the narrowest margins, measured before excision.

Related items that appear with it:

• Multiple lesions are coded separately, each with its own measurement. They are not added together.
• Shave removal, destruction, and excision are different code families. Shaving is a transverse removal without a full-thickness excision; destruction is by an agent — laser, cryotherapy, electrosurgery — with nothing to send to pathology.
• Excision includes simple closure. Reporting a simple repair with an excision is unbundling.

Destruction, shaving, and biopsy

Biopsy removes a portion of a lesion for diagnosis. When a lesion is biopsied and then excised at the same session, the biopsy is generally not separately reported — it is part of getting to the excision. There are exceptions and they are stated in the guidelines.

Shave removal is a transverse or horizontal removal without full-thickness excision, and it does not include a repair, because there is no wound to close in layers.

Destruction — laser, cryosurgery, electrosurgery, chemical — has its own family, and it produces no specimen, which has a downstream consequence: no pathology report, therefore no confirmation of benign versus malignant, therefore the destruction code family has to make that distinction on clinical grounds.

Destruction of premalignant lesions — actinic keratoses and the like — has its own codes and its own counting convention: the first lesion, then an add-on for each additional, up to a point, then a code covering a larger number. Count the lesions and read the descriptors, because the counting convention differs between the premalignant and the benign/malignant destruction families and coders routinely apply one family's rule to the other.

Mohs micrographic surgery

Mohs deserves its own paragraph because it is coded unlike anything else in the section and because it is a documented audit target.

In Mohs micrographic surgery, a single physician acts as both the surgeon and the pathologist, excising tissue in stages and examining each stage's margins microscopically before deciding whether another stage is needed.

That dual role is the definitional requirement. If a different physician performs the pathology, it is not Mohs — it is an excision plus a separately reported pathology service, coded that way.

The codes are structured by anatomic area, by STAGE, and by the number of TISSUE BLOCKS per stage, with add-on codes for additional stages and for additional blocks beyond the first five in a stage. Count stages and blocks from the operative note, not from an impression of how long it took.

And the surgical pathology codes are not reported separately — the microscopic examination is what Mohs is, and reporting it again is duplicate billing. A repair of the resulting defect, however, is separately reportable.

Nails, skin tags, and the small procedures

Three high-volume families that share one characteristic: they are individually inexpensive and collectively substantial, and they are the ones practices most often fail to capture at all.

Nail procedures are coded by what was done — trimming, debridement, avulsion, excision of the nail matrix — and several are coded per nail or in ranges of nails. Read the descriptor for the unit; this family contains both per-nail codes and codes covering multiple nails, and mixing them up produces either a large overcharge or a large undercharge with no middle ground.

Skin tag removal has its own code family with a counting convention — the first several tags, then an add-on for each additional group. It is not an excision and it is not a destruction.

Foreign body removal from subcutaneous tissue is coded separately from the repair of the wound it came out of, and by whether the removal was simple or complicated. "Complicated" requires documentation of what made it so.

17.5 Repairs: simple, intermediate, complex, and adding lengths

Wound repair is coded by three variables, and all three are needed:

1. Complexity — simple, intermediate, or complex
2. Anatomic site — grouped, and the groupings matter
3. Length in centimeters

The three complexities

SIMPLE — superficial; epidermis, dermis, or subcutaneous tissue without significant involvement of deeper structures. One-layer closure. Includes local anesthesia and chemical or electrical cauterization of wounds not closed.

INTERMEDIATE — requires layered closure of one or more of the deeper layers of subcutaneous tissue and superficial fascia, in addition to the skin. Also: a single-layer closure of a heavily contaminated wound that required extensive cleaning or removal of particulate matter.

COMPLEX — more than layered closure. Scar revision, debridement, extensive undermining, stents or retention sutures. Complex repair does not include excision of a benign or malignant lesion.

"Layered closure" is the operative phrase for intermediate, and it is why §17.3's 📋 Read the Chart mattered: the words "closed in layers" in the body of a note are worth a separately reportable code that the procedure heading did not mention.

And the second half of the intermediate definition is the one people miss — a single-layer closure can be intermediate if the wound was heavily contaminated and required extensive cleaning. That has to be documented. "Wound irrigated" is not extensive cleaning; "extensive irrigation and removal of embedded gravel" is.

The anatomic groupings

You cannot apply the adding rule without knowing the groupings, and coders regularly try.

The groupings differ between the three complexity levels, which is the detail that makes this hard and which no summary can rescue you from — you have to read the descriptors in the family you are coding. In general shape:

   SIMPLE repair groups (illustrative shape; read the descriptors)
     · scalp, neck, axillae, external genitalia, trunk, extremities
       (including hands and feet)
     · face, ears, eyelids, nose, lips, mucous membranes

   INTERMEDIATE repair groups
     · scalp, axillae, trunk, extremities (excluding hands and feet)
     · neck, hands, feet, external genitalia
     · face, ears, eyelids, nose, lips, mucous membranes

Notice that hands and feet move between groupings as complexity changes, and that the face group is consistently separate. Two intermediate repairs, one on the trunk and one on the hand, are in different groupings and are reported separately — even though both are intermediate.

That is the single most common repair error, and it is not a rule you can reason your way to. It is a lookup, every time, until it is memorized.

The adding rule

Add together the lengths of repairs in the SAME classification and from the SAME anatomic grouping. Report one code for the sum.

Repairs of different classifications, or from different anatomic groupings, are reported separately.

   THREE LACERATIONS, ONE PATIENT

   left forearm    3.0 cm   intermediate  ┐
   right forearm   2.5 cm   intermediate  ├─ same class, same grouping
                                          │   ► ADD:  5.5 cm, ONE code
   scalp           4.0 cm   simple        ── different class AND grouping
                                              ► REPORT SEPARATELY

Coders get this wrong in both directions — adding across classifications, which inflates, and reporting each laceration separately within a classification, which also inflates. Same class, same grouping, add.

Two rules that prevent unbundling

Simple repair is included in excision codes. Do not report it separately.

Intermediate and complex repairs ARE separately reportable with an excision — which is exactly the money left on the table in §17.3's example.

🔍 Check Your Understanding

A patient sustains four lacerations in one incident:

• 2.0 cm, trunk, layered closure
• 3.5 cm, thigh, layered closure
• 1.5 cm, left hand, layered closure
• 2.0 cm, forehead, single-layer closure

How many repair codes, and what lengths?

Answer: Three.

The trunk (2.0) and the thigh (3.5) are both intermediate and both in the trunk-and-extremities grouping → add: 5.5 cm, one code.

The hand (1.5) is intermediate but sits in a different grouping at this complexity → its own code, 1.5 cm. This is the one people miss.

The forehead (2.0) is a simple repair — a different classification entirely, and a different grouping — → its own code, 2.0 cm.

What would be wrong: adding all four (7.0 cm, one code) — crosses both classification and grouping. Reporting four codes — fails to add the two that belong together. Adding the three intermediate repairs (7.0 cm) — crosses a grouping boundary, which is the subtle error and the common one.

17.6 Grafts, flaps, and debridement depth

Adjacent tissue transfer

Adjacent tissue transfer — Z-plasty, W-plasty, rotation flaps, advancement flaps — is coded by the square centimeters of the defect, which includes both the primary defect and the secondary defect created by moving the tissue.

The excision of the lesion is INCLUDED in the adjacent tissue transfer code. Do not report both.

That is a genuine trap: a lesion excision followed by a rotation flap looks like two procedures in the note and is one code.

Grafts

Skin grafts are coded by type and by recipient site and size:

• Split-thickness versus full-thickness
• Autograft (the patient's own tissue), allograft (donor human tissue), xenograft (non-human)
• Skin substitutes, which have their own family and their own application codes

Size is measured in square centimeters of the RECIPIENT site, and the first 100 sq cm (or 1% of body area for infants and children) is the base, with add-on codes for additional area.

Debridement

Debridement is coded by DEPTH and by AREA, and the depth is the deepest tissue removed:

   DEPTH decides the code family
     skin, subcutaneous tissue
     subcutaneous tissue          ► 11042 family
     muscle and/or fascia         ► deeper family
     bone                         ► deepest family

   AREA decides the units
     first 20 sq cm, then add-on codes for each additional 20 sq cm

Two rules:

The depth is what was REMOVED, not what was exposed. An operative note describing a wound "down to fascia" has described what is visible. Debridement of fascia requires that fascia was debrided.

When multiple wounds are debrided to the same depth, add the areas together. Different depths are reported separately.

And debridement performed as part of another procedure is generally included in it — which is precisely the Account 31-2245 problem, arriving in §17.9.

17.7 The musculoskeletal system and the fracture-care decision

The musculoskeletal section is organized anatomically, head to foot, and within each anatomic area by type of procedure — incision, excision, introduction or removal, repair/revision/reconstruction, fracture and/or dislocation, arthrodesis, amputation.

That organization is the fastest way to navigate it, and it is why coders who work in orthopedics learn the section's shape rather than searching the index every time.

The general musculoskeletal codes, and the joint injection family

The section opens with a general subsection covering procedures not confined to one anatomic area, and it contains the family Account 10-4471 lives in.

Joint and bursa injections and aspirations are coded by the size of the joint and by whether imaging guidance was used:

   JOINT / BURSA ASPIRATION OR INJECTION

     small joint or bursa      (fingers, toes)
     intermediate joint/bursa  (wrist, elbow, ankle, olecranon bursa)
     LARGE joint or bursa      (shoulder, hip, KNEE, subacromial bursa)
         │
         ├── WITHOUT ultrasound guidance ......... 20610
         └── WITH ultrasound guidance,
             with permanent recording and report .. 20611

Three rules govern the whole family.

The joint size is anatomic, not a judgment. A knee is a large joint. Look it up once and you know it forever.

The imaging-guidance code requires more than imaging. 20611 requires permanent recording and report. Ultrasound used to guide a needle, with nothing recorded and nothing reported, does not support it. This is the same principle as Chapter 14 §14.7's professional component: no report, no service.

And the unit is the joint, not the injection. Injecting one knee once and injecting one knee twice in the same session are the same single service. Injecting two different joints is two services — which brings the laterality and bilateral rules of Chapter 14 §14.8 directly into this family.

🗂️ The Encounter — why 20610 and not 20611

The March 14 procedure note, frozen in Chapter 4's Figure 4.2, contains the phrase:

"No imaging guidance used."

Chapter 13 §13.6 flagged that phrase as a documented negative — a sentence that exists to record that something did not happen — and deferred the consequence to this chapter.

This is the consequence. The knee is a large joint. Without imaging guidance, 20610. With ultrasound guidance and a permanent recording and report, 20611, which carries a materially higher value.

The physician wrote one clause and closed a coding question that would otherwise require a query. Without it, a coder faces a note that is silent about guidance, and silence is not a documented negative — it is an absence, and an absence has to be asked about.

This is the case for documented negatives in three sentences, and Chapter 38 §38.1 makes it at length. The note also carries "no aspirate obtained" and "no known injury", each of which decides something: the first that this was an injection rather than an aspiration-and-injection, and the second — as Chapter 12 §12.2 established — which ICD-10-CM chapter the diagnosis comes from.

Three clauses, three questions closed, in a note that took no longer to dictate.

The fracture-care decision

This is a genuine fork, it comes up constantly, and there is no default.

A physician who treats a fracture may report EITHER:

(A) A global fracture care code — which carries a 090-day global period and includes the initial treatment, the first cast or splint, and all normal follow-up including subsequent casting;

(B) An E/M service plus casting/splinting — treating the encounter as an ordinary visit with an immobilization procedure, with no global period.

Both are legitimate. They are not interchangeable, and the choice has consequences.

The decision turns on who is providing the definitive care. An emergency physician who splints a wrist and sends the patient to an orthopedist has not undertaken 90 days of fracture management, and reporting global fracture care would assert that they had. The E/M plus splint is correct.

An orthopedist who accepts that patient and manages them through healing is providing fracture care, and reports the fracture care code.

⚠️ Where Claims Die

Two physicians both reporting global fracture care for the same fracture.

The emergency physician reports fracture care because their system's macro does. The orthopedist reports it because they are, in fact, providing it. One of the two claims will deny, and the argument that follows is unpleasant, because both offices believe they are right and one of them has a 090-day global that will now block ninety days of legitimate follow-up billing.

This is not a coding error in the ordinary sense. It is a coding decision made by a macro, and it is the same failure as Chapter 14 §14.1's automatically appended modifier and Chapter 15's prefilled time: a configuration making an assertion nobody chose to make.

The fix is a rule, written down: in this organization, in this setting, fracture care is reported when these conditions are met. It takes one meeting and it prevents a recurring dispute.

Closed, open, and percutaneous

Fracture treatment codes are further divided by treatment type, which describes what the physician did, not what the fracture was:

• Closed treatment — the fracture site is not surgically opened
• Open treatment — the fracture is surgically opened and visualized, or opened remotely for fixation
• Percutaneous skeletal fixation — neither open nor closed; fixation is placed through the skin under imaging

"Closed treatment" does not mean the fracture was closed rather than compound. It means the physician did not open it. A compound (open) fracture can receive closed treatment, and this single confusion accounts for a large share of the errors in this family.

And with or without manipulation further divides the codes. Manipulation means the fracture was reduced — the fragments were manipulated into position. It must be documented as such.

17.8 Casting, splinting, and who gets paid for it

Three separate things get billed around an immobilization, and they are frequently confused.

The application code. A cast or splint application code is reported when the application is the service — a replacement cast, or an application by a physician who is not reporting fracture care. It is NOT reported when the initial cast or splint is applied as part of global fracture care, because it is included.

The supply. The casting or splinting material itself is reported with a HCPCS Level II Q code — Chapter 20 covers the family. The supply is separately reportable even when the application is included, which surprises people, and it is money practices routinely leave behind.

The E/M, if one is separately reportable, subject to Chapter 14's rules.

   WHO BILLS WHAT

   Global fracture care reported?
     YES → application INCLUDED · supply SEPARATELY REPORTABLE
     NO  → application REPORTABLE · supply REPORTABLE · E/M as applicable

   Replacement cast during a global period?
     If reporting global fracture care ... INCLUDED
     If not .......................... application + supply reportable

📞 On the Phone

"You billed us for the cast material. The visit was included."

A real call, and the caller is half right.

What works: "The application of the cast is included in the fracture care — you're right about that, and we haven't billed for applying it. What's on the claim is the casting material itself, which is a supply and is separately payable. Let me give you the line and the code so you can see the two are different."

What does not work: conceding it. It is a correct charge and writing it off to end an uncomfortable call trains everyone involved — including your own billing office — that it was wrong.

And know the exception before you make the call: a payer may have its own policy bundling the supply, and some contracts do. Check the policy first, so that if you are wrong you find out before the patient does.

17.9 Arthroscopy and the shoulder that started an audit

The rule

Surgical arthroscopy always includes diagnostic arthroscopy.

Every time. A surgeon who looks and then repairs has performed one arthroscopic procedure, not two, and reporting a diagnostic arthroscopy alongside a surgical one is unbundling in its simplest form.

The knee, and the compartment convention

The knee arthroscopy family is where most coders meet this section, and it is organized around a concept that does not appear anywhere else in CPT.

The knee has three compartments — medial, lateral, and patellofemoral — and several knee arthroscopy codes are written in terms of which compartment was worked in.

Two consequences.

Procedures in different compartments may be separately reportable, where the codes and the edits permit it — and the operative note must name the compartment. A note saying "debridement performed" has not established a compartment. A note saying "debridement of the lateral compartment" has.

Procedures in the same compartment generally are not. A meniscectomy and a debridement in the same compartment are one service, and appending a modifier to make them two is Account 31-2245 with a different joint.

This is the same discipline as §17.6's debridement depth and §17.9's shoulder: the code depends on an anatomic fact, and the anatomic fact has to be in the note.

The "separate procedure" designation, applied

Chapter 13 §13.8 introduced the "separate procedure" parenthetical designation and stated the rule. The surgery sections are where it bites.

A code designated "(separate procedure)" is reportable only when it was performed alone, or when it was independent of and unrelated to other services performed at the same session. When it is a component of a larger procedure, it is not reported.

Diagnostic arthroscopy is the canonical example and it is why the rule at the top of this section is not an arbitrary bundling edit — it follows directly from the designation.

Read the parenthetical. It is three words in italics and it changes whether a code may appear on the claim at all.

The shoulder — Account 31-2245

Chapter 5 introduced this file. It is time to read the claim as a coder.

The claim as submitted (constructed; Ridgeview Orthopedic Surgery):

Line 1 — 29827, arthroscopic rotator cuff repair. The primary procedure. 090-day global.

Line 2 — 29826. This is an add-on code, and Chapter 13 §13.7's rules apply: it cannot be reported alone, modifier 51 does not belong on it, and it is not subject to the multiple-procedure reduction. Its global indicator is ZZZ — it lives inside 29827's global period. Line 2 is correct.

Line 3 — 29822 with modifier 59. This is the problem.

29822 is limited debridement. Under NCCI it is bundled into 29827 when the debridement is in the same anatomic region as the repair, and the operative note documents debridement of the same structures that were repaired.

Modifier 59 was appended by a billing macro, not by a coder reading an operative report.

Chapter 14 §14.5 stated the discipline: before appending modifier 59, find the sentence in the operative note. Here there is no such sentence. There is a sentence saying the opposite.

What it cost

(Constructed.) A commercial payer's special investigations unit reviewed 42 claims over 18 months with the identical pattern, found an average overpayment of \$612.40 per claim, and demanded:

   42 claims  ×  $612.40  =  $25,720.80

And then the part that matters most. The practice's own subsequent internal review found that 11 of the 42 were, in fact, separately documented and defensible.

They could not prove it.

The debridement in those eleven cases had been performed in a different anatomic region, or on different structures, and the operative notes did not say so with enough specificity to establish it after the fact. The surgeons remembered. Memory is not documentation.

⚖️ Compliance Check

The defense is contemporaneous documentation, and it cannot be built retroactively.

This is the most important sentence in Part III and it is why Account 31-2245 exists in this book.

Eleven claims out of forty-two were probably right. Eleven claims out of forty-two were paid back, because a note that says "debridement performed" does not establish a distinct anatomic region and nothing written eighteen months later can make it say so.

What would have prevented it is not a compliance program, an audit, or a policy. It is a sentence in an operative note: "Debridement was performed in the subacromial space, anatomically distinct from the repaired supraspinatus insertion." One sentence, dictated at the time, is the entire difference between \$25,720.80 and \$18,984.40.

And the macro that appended modifier 59 to every 29822 is the reason nobody ever looked for that sentence — because with the macro running, nobody had to read the note at all.

17.10 Coding a surgery from the note, twice

The chapter's method, applied end to end. Read the note, code it, then read it again looking specifically for what you missed the first time.

The note (constructed):

Preoperative diagnosis: Lesion, left forearm. Postoperative diagnosis: Same, pending pathology. Procedure performed: Excision of lesion, left forearm.

Body: "…the lesion measured 1.8 cm in greatest dimension. It was excised with 0.4 cm margins circumferentially, and the specimen was submitted to pathology. The resulting defect measured 2.6 cm. Hemostasis was obtained. The deep dermal layer was approximated with interrupted 4-0 absorbable suture and the skin was closed with running 5-0 nylon. Sterile dressing applied."

First pass

An excision of a lesion. Size: 1.8 + 0.4 + 0.4 = 2.6 cm. Anatomic site: forearm — an upper extremity grouping. Benign or malignant: unknown, pending pathology — so the code cannot be finalized today.

A coder who stops here reports one code and is wrong.

Second pass — what the first pass missed

The closure. "The deep dermal layer was approximated…and the skin was closed" is a layered closure, which is an intermediate repair, which is separately reportable with an excision — §17.5.

Its measurement is the DEFECT, 2.6 cm, not the excision. The two are the same number here by coincidence.

The heading did not mention it. §17.3's rule: the body governs.

The claim

And what would have made it wrong

• Coding 1.8 cm — the lesion without margins
• Coding 2.3 cm — the formalin-shrunken pathology specimen
• Reporting a simple repair — included in the excision; unbundling
• Reporting only the heading's procedure — leaving the repair unbilled
• Coding it today rather than holding for pathology

Five ways to be wrong on a two-line claim, and four of them come from reading one number or one heading instead of the note.

🔍 Check Your Understanding

The pathology returns malignant, and the surgeon documents that a re-excision was performed at a later date to obtain clear margins.

Which global period modifier applies to the re-excision, and why?

Answer: Modifier 58 — a staged or related procedure during the postoperative period. The re-excision is planned in the sense that matters: it is a related procedure undertaken because the first one's result required it, not an unplanned return to the operating room for a complication. 58, not 78 — Chapter 14 §14.9. And 58 generally pays in full and starts a new global period, which is the correct outcome for a service that is more definitive treatment rather than the repair of a problem.

Summary

The surgical package includes the E/M after the decision for surgery, local anesthesia, immediate postoperative care, orders, recovery-room evaluation, and typical follow-up. It excludes the visit at which the decision for surgery was made (modifier 57), complications requiring a return to the operating room (78), unrelated services (79, 24), treatment of the underlying condition, and supplies beyond the usual. The decision-for-surgery visit is the largest silent loss in surgical billing.

Global periods: 000 minor, no postoperative days · 010 minor, 10 days · 090 major, one preoperative day plus 90 · XXX not applicable · YYY payer-determined · ZZZ inside another code's global (add-on codes). Minor versus major is a lookup, not a judgment, and it decides modifier 25 versus 57. The Medicare Physician Fee Schedule relative value file carries it, free.

Read an operative report in order: postoperative diagnosis, then the body. The "Procedure Performed" heading is a summary and is not authoritative. When it disagrees with the body, the body wins and you query.

THE MEASUREMENT RULE

Lesion's greatest diameter + the narrowest required margins, measured BEFORE excision. Not the pathology specimen (formalin shrinks tissue). Not the defect (that is the repair's measurement).

Repairs: simple (one layer, included in excision), intermediate (layered closure, or a single layer of a heavily contaminated wound requiring extensive cleaning), complex (more than layered). Add lengths within the same classification and anatomic grouping; report different classifications separately. Intermediate and complex ARE separately reportable with an excision.

Adjacent tissue transfer includes the lesion excision. Grafts are coded by type and recipient site area. Debridement is coded by depth removed — not exposed — and by area, with same-depth areas added.

Mohs requires that ONE physician be both surgeon and pathologist. If someone else reads the pathology, it is not Mohs. Coded by area, stage, and tissue blocks; the pathology is not separately reported; the repair is.

Joint injections and aspirations are coded by joint size and by imaging guidance — 20610 without ultrasound, 20611 with it plus a permanent recording and report. The unit is the joint, not the injection.

The knee has three compartments, and several arthroscopy codes are written in terms of them. The compartment has to be named in the note. And a code marked "(separate procedure)" is reportable only when performed alone or independently — which is where the diagnostic-arthroscopy rule comes from.

The fracture-care decision: global fracture care (090 global, follow-up and cast changes included) or an E/M plus casting (no global, follow-up billable). The fork turns on who is providing definitive care. "Closed treatment" means the physician did not open it — not that the fracture was not compound.

Casting: the application is included in global fracture care; the supply is separately reportable either way.

Surgical arthroscopy always includes diagnostic arthroscopy. Account 31-2245's line 3 — 29822 with modifier 59 appended by a macro against a note documenting debridement of the repaired structures — produced a \$25,720.80 demand across 42 claims. Eleven were probably defensible and could not be proven. The defense is contemporaneous documentation and it cannot be built retroactively.

Account 10-4471's 20610 carries a 000-day global, which includes the local anesthetic and the immediate pre- and post-procedure work, excludes the separately identifiable E/M and the drug, and carries no postoperative days at all.

Key Terms

Surgical package — the set of services always included in a surgical procedure's payment, in addition to the operation. (Ch.17)

Global period — the duration of the surgical package, expressed as an indicator: 000, 010, 090, XXX, YYY, or ZZZ. (Ch.17)

Preoperative component — for a 090-day global, the day before surgery, included in the package. (Ch.17)

Intraoperative component — the operation itself. (Ch.17)

Postoperative component — typical follow-up care through the end of the global period, included in the package. (Ch.17)

Staged procedure — a planned or related subsequent procedure during a postoperative period, reported with modifier 58. (Ch.17)

Lesion excision measurement — the lesion's greatest diameter plus the narrowest required margins, measured before excision. Not the pathology specimen and not the defect. (Ch.17)

Margins — the rim of normal tissue excised around a lesion; counted on both sides in the excision measurement. (Ch.17)

Simple repair — one-layer closure of a superficial wound. Included in excision codes. (Ch.17)

Intermediate repair — layered closure involving deeper subcutaneous tissue or superficial fascia in addition to skin; or single-layer closure of a heavily contaminated wound requiring extensive cleaning. Separately reportable with an excision. (Ch.17)

Complex repair — repair requiring more than layered closure, such as scar revision, extensive undermining, or retention sutures. (Ch.17)

Adjacent tissue transfer — the movement of adjacent tissue to close a defect, coded by the square centimeters of primary plus secondary defect. Includes the lesion excision. (Ch.17)

Debridement depth — the deepest tissue actually removed, which determines the code family. What was exposed is not what was debrided. (Ch.17)

Fracture care — global management of a fracture, carrying a 090-day global period and including initial treatment, the first cast, and normal follow-up. (Ch.17)

Closed treatment — treatment in which the fracture site is not surgically opened. Not a statement about whether the fracture was compound. (Ch.17)

Open treatment — treatment in which the fracture is surgically opened and visualized, or opened remotely for fixation. (Ch.17)

Manipulation — reduction of a fracture; the fragments were manipulated into position. Must be documented. (Ch.17)

Casting and splinting supplies — the materials, reported with HCPCS Level II Q codes and separately reportable even when the application is included. (Ch.17)

Mohs micrographic surgery — staged excision with microscopic margin examination in which a single physician acts as both surgeon and pathologist. Coded by anatomic area, stage, and tissue blocks; the pathology is not separately reported. (Ch.17)

Joint size (injection family) — the anatomic classification — small, intermediate, or large — that selects the arthrocentesis or injection code. A knee is a large joint. (Ch.17)

Knee compartments — medial, lateral, and patellofemoral; the anatomic divisions several knee arthroscopy codes are written in terms of, which must be named in the operative note. (Ch.17)

"Separate procedure" designation — a parenthetical marking a code reportable only when performed alone or independently of other services at the same session. (Ch.17)

Spaced Review

1. Name three things included in the surgical package and three things excluded. Which excluded item is most often written off, and why does nobody notice?

2. A code carries a ZZZ global indicator. What kind of code is it, and what does Chapter 13 §13.7 forbid you to do with it?

3. (Chapter 14) A physician performs a procedure with a 090-day global and reports an E/M on the same date at which the decision to operate was made. Which modifier, and why not 25?

4. State the lesion excision measurement rule. Then say why the pathology report is the wrong source and in which direction the error runs.

5. Three lacerations: 3.0 cm and 2.5 cm intermediate repairs of the forearms, and a 4.0 cm simple repair of the scalp. How many repair codes, and what lengths?

6. A patient arrives at an emergency department with a wrist fracture. The physician splints it and refers to orthopedics. Which fracture-care option is correct, and what would the other one assert?

7. (Chapter 5) Account 31-2245's line 3 carried modifier 59 from a macro. Eleven of the forty-two claims were probably defensible. State, in one sentence, why the practice paid for all forty-two.