Actinobacillus pleuropneumoniae
LEVELS: Highly unlikely; No human illness; Routine biosecurity keeps it out; Routine; Moderate; Negligible; Some; Available but inconsistent
| Register id | actinobacillus_app |
| Type | bacteria |
| Scientific name | Actinobacillus pleuropneumoniae |
| NCBI taxid | 715 |
| Evidence | 1 document(s) |
| Assigned | 2026-09-04, against criteria version 093366e352a2 |
Overview
Actinobacillus pleuropneumoniae, usually shortened to APP, is a small gram-negative bacterium that causes porcine pleuropneumonia — one of the most important bacterial respiratory diseases of pigs worldwide. It lives in the tonsils of carrier pigs, and many conventional herds carry it without ever showing disease: 78% of Canadian herds with no clinical pleuropneumonia still tested positive. When disease does break out it can be dramatic. The most virulent strains produce Apx toxins that destroy lung tissue, and pigs of any age can die of a haemorrhagic, necrotising pneumonia within hours of first appearing sick. Survivors are left with scarred lungs and grow poorly. Nineteen serotypes are recognised and they differ markedly in virulence, which matters practically because vaccines are largely serotype-specific. Spread between herds is almost always by moving an infected but healthy-looking pig; within a herd it passes by nose- to-nose contact and short-range droplets. High-mortality outbreaks have become uncommon in the United States and Canada but remain a serious problem in Latin America, Asia and Europe, and a 2023 serotype 15 outbreak killed a quarter of the pigs on several US finishing sites.
C1 Zoonotic potential
Likelihood of transmission from pigs to humans
Highly unlikely: Human infection with the agent has never been reported, or has been reported but is not attributed to pig or pork exposure
Basis. ch43 43.1.3, a Public Health section of its own: "APP DOES NOT USUALLY INFECT HUMANS AND POSES NO PUBLIC HEALTH RISK. However, necrosis from needlestick injury with live APP vaccine (non-virulent strain for pigs) (Rycroft et al. 2011) and THE ISOLATION OF AN APP SEROTYPE 5 STRAIN FROM A POORLY HEALING WOUND OF A SWINE PRODUCER AFTER BEING BITTEN BY A BOAR (unpublished observations) have been reported." 43.1.4.1: "APP is widely distributed and ONLY INFECTS PIGS, although a report suggested that an APP or APP-like strain was able to infect poultry."
How this level was decided
Level 1, and it is the worked example at this level in criteria_levels.yml. TWO ANOMALIES ARE ON RECORD AND ARE STATED HERE RATHER THAN SUPPRESSED, because a reader who finds them later will think they were missed: a needlestick with a live vaccine strain, which is inoculation trauma rather than transmission, and one unpublished isolation from a boar-bite wound, which is a bite wound contaminated with the biter oral flora. Neither is natural transmission of infection from pigs to a person, and the summary sentence of the chapter -- no public health risk, only infects pigs -- is written knowing both. Level 1 stands.
C2 Zoonotic impact
Severity of human illness caused by the agent
No human illness: Agent does not cause illness in people
Basis. ch43 43.1.3: "APP does not usually infect humans and POSES NO PUBLIC HEALTH RISK." No human illness with APP is described anywhere in the chapter.
How this level was decided
Level 1, the worked example at this level. Under C2 as rescoped on 2026-09-01, level 1 means pigs are not a source of human illness with this agent, which is exactly what the chapter says. Follows C1.
C3 Herd introduction risk
Effort required to keep the agent out of the herd
Routine biosecurity keeps it out: Quarantine of incoming stock, transport and fomite control, cleaning and disinfection, and the usual monitoring reliably exclude it
Basis. ch43 43.1.11: "When herds are free of APP, STRICT BIOSECURITY MEASURES should be applied to prevent introduction of the organism. THE GREATEST RISK IS PRESENTED BY THE INTRODUCTION OF CLINICALLY HEALTHY INFECTED PIGS. Seedstock should be purchased from herds with a history of absence of clinical sign and APP-compatible lesions, and SEROLOGICAL RESULTS MUST BE NEGATIVE for all known APP serotypes if animals are to be introduced in an APP-free herd. IDEALLY, INCOMING ANIMALS SHOULD BE QUARANTINED AND ALSO SEROLOGICALLY TESTED BEFORE INTRODUCTION." 43.1.4.2 on how it moves: "TRANSMISSION BETWEEN HERDS OCCURS MAINLY THROUGH THE INTRODUCTION OF CARRIER ANIMALS. The main route of spread is by direct nose-to-nose contact or by droplets within short distances." On fomites, the 2023 US serotype 15 outbreak: "the more credible explanation for the outbreaks was A SERIOUS GAP IN BIOSECURITY PROCEDURES, especially concerning activities of the rendering company, WHICH SHARED RENDERING TRUCKS, ROUTES, AND DRIVERS dedicated to the area." On aerosol between herds: "Kristensen et al. (2004b) reported that AIRBORNE TRANSMISSION BETWEEN CLOSELY LOCATED PIG UNITS IS POSSIBLE, BUT UNCOMMON. In contrast, Zhuang et al. (2007) suggested that local spread of APP serotype 2 from neighboring herds was a predominant factor in the contamination of Danish specific pathogen-free (SPF) genetic herds." 43.1.11.1 records that eradication succeeds: depopulation and restocking, or off-site segregated medicated early weaning supported by vaccination, medication and culling.
How this level was decided
Level 3, and it is the worked example at this level. The chapter prescribes the level 3 list almost item for item -- quarantine of incoming stock, serological monitoring, fomite and transport control -- and names carrier pigs as the main route, which is the route housing does nothing about and routine biosecurity does. THE ARGUMENT FOR LEVEL 4 IS REAL AND IS SET OUT HERE RATHER THAN IGNORED: carrier detection is genuinely hard (43.1.9.2 calls selective culture of tonsils very low in sensitivity and immunomagnetic separation time-consuming and expensive), and the level 4 label names "repeated costly testing to find carriers"; and two studies report herd-to-herd aerosol spread, which is what air filtration is for. IT DOES NOT CARRY, for two reasons. Serology is the practical exclusion tool and the chapter calls it "the preferred and most cost-effective method for APP surveillance" -- that is routine monitoring, not extraordinary testing. And APP-free herds are in fact established and held in North America by source-herd selection, quarantine and serology, without anyone filtering air for it; the herds that filter do so for PRRSV. Herd-to-herd aerosol is reported as possible but uncommon by the one study that measured it directly.
C4 Detection difficulty
Difficulty of recognizing and confirming infection
Routine: The disease would be suspected from clinical signs, history and epidemiology in normal practice, and confirmed by tests that local or regional laboratories run routinely
Basis. ch43 43.1.9.1: "APP PLEUROPNEUMONIA SHOULD BE SUSPECTED WHEN TYPICAL CLINICAL SIGNS AND GROSS LESIONS ARE OBSERVED... Diagnosis is confirmed by culture, identification, and often typing of APP... IT IS RELATIVELY EASY TO DEMONSTRATE APP BY CULTURE IN PNEUMONIC LESIONS FROM FRESHLY DEAD ANIMALS. Primary isolation... may be carried out on 5% sheep blood agar with a cross-streak of S. aureus or S. epidermidis... THIS ALLOWS A RAPID PRESUMPTIVE BACTERIOLOGIC DIAGNOSIS." MALDI-TOF MS "can be used for routine identification of isolates", and direct PCR on lung imprints made on FTA cards detects and even serotypes. 43.1.9.3: "SEROLOGY IS THE PREFERRED AND MOST COST-EFFECTIVE METHOD FOR APP SURVEILLANCE... COMMERCIAL KITS EXIST for both types of ELISA."
How this level was decided
Level 1, the worked example at this level. Both halves of the label are met: the clinical picture and the fibrinohemorrhagic necrotising pleuropneumonia with interlobular fibrin point at this disease in normal practice, and confirmation is ordinary bacteriology plus commercial ELISA at any regional diagnostic laboratory. NOTE THE SCOPE OF THE LEVEL, because the chapter spends five pages on the harder problem: CONFIRMING CLINICAL DISEASE IS ROUTINE, DETECTING SUBCLINICAL TONSILLAR CARRIERS IS NOT. C4 asks about recognizing and confirming infection in the animal that is sick; the carrier-detection problem is scored where it belongs, in C3.
C5 Production cost
Financial impact on the infected farm's cost of production
Moderate: Losses measurably increase cost of production but remain manageable within normal farm operations, whether acute, chronic, or associated with endemic disease
Basis. ch43 43.1.1: "The economic importance of APP is mainly due to the MORTALITY, REDUCED GROWTH, VETERINARY COSTS (ANTIMICROBIALS, VACCINATIONS), AND CONDEMNATIONS AT THE SLAUGHTERHOUSE. However, in chronically infected herds, RESULTS FROM STUDIES INVESTIGATING IMPACT ON AVERAGE DAILY GAIN HAVE BEEN CONTROVERSIAL... HIGH MORTALITY OUTBREAKS ARE NOW RATHER INFREQUENT IN THE UNITED STATES AND CANADA, but remain a problem in Latin American, Asian, and European countries... APP REMAINS A SIGNIFICANT CAUSE OF ECONOMIC LOSS TO THE SWINE INDUSTRY." The recent US event, 43.1.4.1: "a serious outbreak with HIGH MORTALITY (25%) due to this serotype has been reported in SEVERAL WEAN-TO-MARKET OR FINISHER SITES IN THE UNITED STATES (Machado et al. 2023)." On the endemic baseline, 43.1.4.1: 78% of Canadian herds free of clinical pleuropneumonia were APP positive by PCR.
How this level was decided
Level 3. Scored on modern US production per the standing note SCORING/production_cost_impact, and the chapter makes that easy by drawing the geographic line itself: high-mortality outbreaks are infrequent in the US and Canada, while most conventional herds carry low-virulence serotypes with no clinical disease at all. What remains is real and continuous -- vaccination and antimicrobial cost, chronic pleuritis and slaughterhouse condemnation, and the contribution to the porcine respiratory disease complex -- and it is absorbed within normal farm operations, which is the level 3 label in terms. IT IS NOT LEVEL 4: the 2023 serotype 15 outbreak shows the tail is severe on the affected site, but level 4 is a major and potentially unsustainable increase in cost of production, and APP is a disease the US industry has been living with, vaccinating against and eradicating from individual herds for decades.
C6 Market impact
Duration of material disruption to pork and pig markets
Negligible: Little or no material disruption to supply or demand when disease occurs on one or more farms
Basis. APP is endemic in US herds -- 78% of clinically free Canadian herds were PCR positive (43.1.4.1) -- and no trade measure, movement restriction or consumer response following an APP diagnosis is recorded anywhere in chapter 43. The 2023 US serotype 15 outbreak across several sites (Machado et al. 2023) is discussed entirely as a biosecurity failure, with no market consequence mentioned.
How this level was decided
Level 1, and it is the worked example at this level. This is the scoreable case under the standing note SCORING/market_impact rather than a void: APP has been diagnosed in US pigs for decades, including a multi-site high-mortality outbreak in 2023, and no market move has ever followed. The absence is an observation, not an assumption.
C7 Treatment potential
Potential for treatment to improve outcomes
Some: Treatment exists but is inconsistent, requires extralabel use, or works only in some circumstances
Basis. ch43 43.1.10: "ANTIBIOTIC THERAPY IS EFFECTIVE ONLY IN THE INITIAL PHASE OF THE DISEASE when it can reduce mortality (Desrosiers 2004)... The success of therapy depends mainly on early detection of clinical signs and on rapid intervention." On consistency: "RESPONSE OF PIGS WITH CLINICAL APP TO PENICILLIN TREATMENT MAY BE INCONSISTENT (Sjolund et al. 2009)", and antibiotics "significantly varied in their capacity to control acute infection." On what treatment does not achieve: "In spite of apparent clinical success, IT MUST BE REMEMBERED THAT ANTIBIOTIC THERAPY DOES NOT ELIMINATE INFECTION IN A HERD and carrier animal may persist for a long time... SEVERELY AFFECTED ANIMALS MAY NOT RECOVER EVEN AFTER TREATMENT AND SHOULD BE EUTHANIZED." Susceptibility is otherwise good -- ceftiofur, florfenicol, tilmicosin, tiamulin, enrofloxacin and tulathromycin are all reported effective -- but "a high level of resistance to chlortetracycline (88.4%) and oxytetracycline (90.7%) was observed (Archambault et al. 2012). Unfortunately, there is no recent data available."
How this level was decided
Level 2, the worked example at this level, and this entry is the clean illustration of why THE LABEL TEST DOES NOT MECHANICALLY FORCE LEVEL 1. US labels do plainly name Actinobacillus pleuropneumoniae -- it is one of the few swine bacteria named on multiple respiratory product labels -- so the first half of the 2026-09-02 rule is satisfied. The level is still 2 because the other clause of the level 2 label is satisfied too: treatment "works only in some circumstances". The chapter says so three separate ways -- effective only in the initial phase, response inconsistent, severely affected animals may not recover and should be euthanized -- and adds that treatment never clears the carrier state, so it does not solve the herd problem even when it saves the pig. Level 1 would assert that animals recover acceptably, and the pig left with an infarcted, sequestrum-bearing lobe does not.
C8 Vaccine availability
Availability of effective vaccines or bacterins
Available but inconsistent: Commercial or autogenous vaccines exist in the US but protection may be inconsistent
Basis. ch43 43.1.11: "A WIDE RANGE OF VACCINES HAVE BEEN DEVELOPED for this disease... Commercial vaccines fall into three main groups: KILLED ORGANISMS (BACTERINS), SUBUNIT TOXIN-BASED VACCINES, AND TOXIN-EXPRESSING BACTERINS." On how well they work: "THESE VACCINES CAN BASICALLY REDUCE CLINICAL SIGNS, TREATMENTS, AND LUNG LESIONS (Del Pozo et al. 2014), BUT THEY ARE POORLY EFFECTIVE IN ELIMINATING THE CARRIER STAGE." On predictability: "Vaccination with bacterins is SEROTYPE-SPECIFIC; possible cross-immunity with cross-reacting serotypes has been suggested, ALTHOUGH IT HAS NOT BEEN CONFIRMED. The use of a bacterin should be done only in farms where the involved serotype has been identified." And: "IT IS EXTREMELY HARD TO PREDICT WHICH TYPE OF VACCINE IS WORKING BETTER (toxins, bacterins, bacterins + toxins) IN A GIVEN SITUATION, and this seems to depend on the farm and the strain involved." Autogenous vaccines are used in some countries where the clinical serotype is not in a commercial bacterin.
How this level was decided
Level 2. Vaccines are commercially available in the US in three distinct technologies plus autogenous product, so this is not level 3, and the disease plainly justifies vaccinating, so it is not the "not needed" half of level 1. What holds it at level 2 is stated by the chapter in terms: bacterin protection is serotype-specific and requires knowing the serotype on the farm first, cross-protection is unconfirmed, no product clears the carrier state, and which technology works on a given farm cannot be predicted. That is the level 2 label -- available, but protection may be inconsistent.
Levels and evidence are generated from data/assignments/actinobacillus_app.yml; the overview is authored in data/overviews/actinobacillus_app.md. Do not hand-edit this page — corrections go in data/assignments/CORRECTIONS.yml.
Quoted material marked Basis is from Zimmerman JJ, Karriker LA, Ramirez A, Schwartz KJ, Stevenson GW, Zhang J, eds. Diseases of Swine, 12th edition. Hoboken: Wiley Blackwell, 2025 — except where another source is named in the quotation itself.